[House Hearing, 116 Congress]
[From the U.S. Government Publishing Office]
ABOUT FACE: EXAMINING THE DEPARTMENT OF HOMELAND SECURITY'S USE OF
FACIAL RECOGNITION AND OTHER BIOMETRIC TECHNOLOGIES
=======================================================================
HEARING
before the
COMMITTEE ON HOMELAND SECURITY
HOUSE OF REPRESENTATIVES
ONE HUNDRED SIXTEENTH CONGRESS
FIRST SESSION
__________
JULY 10, 2019
__________
Serial No. 116-31
__________
Printed for the use of the Committee on Homeland Security
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Available via the World Wide Web: http://www.govinfo.gov
__________
U.S. GOVERNMENT PUBLISHING OFFICE
38-784 WASHINGTON : 2020
COMMITTEE ON HOMELAND SECURITY
Bennie G. Thompson, Mississippi, Chairman
Sheila Jackson Lee, Texas Mike Rogers, Alabama
James R. Langevin, Rhode Island Peter T. King, New York
Cedric L. Richmond, Louisiana Michael T. McCaul, Texas
Donald M. Payne, Jr., New Jersey John Katko, New York
Kathleen M. Rice, New York John Ratcliffe, Texas
J. Luis Correa, California Mark Walker, North Carolina
Xochitl Torres Small, New Mexico Clay Higgins, Louisiana
Max Rose, New York Debbie Lesko, Arizona
Lauren Underwood, Illinois Mark Green, Tennessee
Elissa Slotkin, Michigan Van Taylor, Texas
Emanuel Cleaver, Missouri John Joyce, Pennsylvania
Al Green, Texas Dan Crenshaw, Texas
Yvette D. Clarke, New York Michael Guest, Mississippi
Dina Titus, Nevada
Bonnie Watson Coleman, New Jersey
Nanette Diaz Barragan, California
Val Butler Demings, Florida
Hope Goins, Staff Director
Chris Vieson, Minority Staff Director
C O N T E N T S
----------
Page
Statements
The Honorable Bennie G. Thompson, a Representative in Congress
From the State of Mississippi, and Chairman, Committee on
Homeland Security:
Oral Statement................................................. 1
Prepared Statement............................................. 3
The Honorable Mike Rogers, a Representative in Congress From the
State of North Carolina, and Ranking Member, Committee on
Homeland Security:
Oral Statement................................................. 16
Prepared Statement............................................. 18
Witnesses
Mr. John P. Wagner, Deputy Executive Assistant Commissioner,
Office of Field Operations, U.S. Customs and Border Protection,
U.S. Department of Homeland Security:
Oral Statement................................................. 19
Prepared Statement............................................. 21
Mr. Austin Gould, Assistant Administrator for Requirements and
Capabilities Analysis, Transportation Security Administration,
U.S. Department of Homeland Security:
Oral Statement................................................. 28
Prepared Statement............................................. 32
Mr. Joseph R. Di Pietro, Chief Technology Officer, U.S. Secret
Service, U.S. Department of Homeland Security:
Oral Statement................................................. 37
Prepared Statement............................................. 38
Mr. Charles H. Romine, Ph.D., Director of Information Technology
Laboratory, National Institute of Standards and Technology,
U.S. Department of Commerce:
Oral Statement................................................. 41
Prepared Statement............................................. 42
For the Record
The Honorable Bennie G. Thompson, a Representative in Congress
From the State of Mississippi, and Chairman, Committee on
Homeland Security:
Article, Washington Post, June 10, 2019........................ 4
Article, Washington Post, July 7, 2019......................... 5
Letter From Todd Hauptli, American Association of Airport
Executives................................................... 8
Statement of the International Biometrics + Identity
Association.................................................. 9
Letter......................................................... 12
The Honorable Mike Rogers, a Representative in Congress From the
State of North Carolina, and Ranking Member, Committee on
Homeland Security:
Letter From Don Erikson to Chairman Bennie G. Thompson and
Ranking Member Mike D. Rogers................................ 49
The Honorable Clay Higgins, a Representative in Congress From the
State of Louisiana:
Article, New York Times, June 9, 2019.......................... 69
Article, TheHill.com, May 9, 2019.............................. 71
The Honorable Debbie Lesko, a Representative in Congress From the
State of Arizona:
Letter From Sharon Pinkerton to Chairman Bennie G. Thompson and
Ranking Member Mike Rogers................................... 72
Letter From Douglas E. Lavin to Chairman Bennie Thompson and
Ranking Member Mike Rogers................................... 73
Letter From Shane C. Downey to Ranking Member Mike Rogers...... 74
The Honorable Sheila Jackson Lee, a Representative in Congress
From the State of Texas:
Article, Houston Chronicle, July 5, 2019....................... 77
Article, CNET, July 8, 2019.................................... 78
Article, New York Times, July 7, 2019.......................... 79
Article, New York Times, July 26, 2018......................... 82
ABOUT FACE: EXAMINING THE DEPARTMENT OF HOMELAND SECURITY'S USE OF
FACIAL RECOGNITION AND OTHER BIOMETRIC TECHNOLOGIES
----------
Wednesday, July 10, 2019
U.S. House of Representatives,
Committee on Homeland Security,
Washington, DC.
The committee met, pursuant to notice, at 10:01 a.m., in
room 310, Cannon House Office Building, Hon. Bennie G. Thompson
(Chairman of the committee) presiding.
Present: Representatives Thompson, Jackson Lee, Langevin,
Richmond, Payne, Correa, Torres Small, Rose, Underwood,
Slotkin, Cleaver, Green of Texas, Clarke, Watson Coleman,
Barragan, Demings, Rogers, McCaul, Katko, Walker, Higgins,
Lesko, Green of Tennessee, Taylor, Joyce, and Guest.
Chairman Thompson. The Committee on Homeland Security will
come to order. The committee is meeting today to receive
testimony on the Department of Homeland Security's use of
facial recognition and other biometric technologies.
Without objection, the Chair is authorized to declare the
committee in recess at any point.
I now recognize myself for an opening statement.
Good morning. The Committee on Homeland Security is meeting
to examine the Department of Homeland Security's use of facial
recognition and other biometric technologies. The Government's
use of biometrics is not entirely new. For example,
fingerprints have been used as an identification tool for many
decades. Other biometrics include DNA, voice pattern, and palm
prints. In recent years, facial recognition has become the new
chosen form of biometric technology.
As facial recognition technology has advanced, its use by
the Government and the private sector has also increased.
Currently, DHS is collecting and storing several different
kinds of biometric information and is using this information
for multiple purposes. CBP and TSA are using biometrics to
conform the identities of travelers, for example. The Secret
Service is piloting a surveillance system using facial
recognition.
I am not opposed to biometric technology and recognize it
can be valuable to Homeland Security in facilitation. However,
its proliferation across DHS raises serious questions about
privacy, data security, transparency, and accuracy. The
American people deserve answers to those questions before the
Federal Government rushes to deploy biometrics further.
Last month, the committee held roundtable discussions with
both industry and privacy and civil liberty stakeholders about
the Department of Homeland Security's increasing use of
biometric technology. Stakeholders have sufficient concerns
that the data DHS is collecting and whether the Department is
safeguarding our rights appropriately. They have good reasons
to be concerned.
Absent standards, Americans may not know when, where, or
why the Department is collecting their biometrics. People also
may not know that they have the right to opt out or how to do
so. Worse yet, they may not know that biometric technology is
in use as it is the case when face recognition is used to
passively surveil a crowd like under the Secret Service's pilot
program.
Recent reports also indicate ICE has been scanning through
millions of Americans' driver's license photos without their
knowledge or consent. These troubling reports are a stark
reminder that biometric technologies should only be used for
authorized purposes in a fully transparent manner.
Data security is another important concern. Frankly, the
Federal Government does not have a great track record securing
Americans' personal data, and biometric information can be
particularly insensitive. Unfortunately, earlier this year, a
CBP subcontractor experienced a significant data breach,
including travelers' images, raising important questions about
data security.
Americans want to know that, if the Government collects
their biometric data, they are going to keep it secure from
hackers and other bad actors. Moreover, the accuracy of certain
biometric technology is in question. Despite advancement in
recent years, studies by highly-regarded academic institutions
have found facial recognition systems in particular are not as
accurate for women and darker-skinned individuals.
Last July, the American Civil Liberties Union conducted a
test using Amazon's facial recognition tool, called
Rekognition. ACLU built a database of 25,000 publicly-available
arrest photos. Using Rekognition, the ACLU searched the
database using pictures of every current Member of Congress.
That software incorrectly matched 28 Members with individuals
who had criminal records. Although the misidentified Members
included both Democrats and Republicans, men and women, and a
wide range of ages, nearly 40 percent of the false matches were
people of color. This is unacceptable. It is not fair to expect
certain people in our society to shoulder a disproportionate
burden of the technology's shortcoming. Before the Government
deploys these technologies further, they must be scrutinized,
and the American public needs to be given a chance to weigh in.
Biometrics and facial recognition technology may be a
useful Homeland Security and facilitation tool, but as with any
tool, it has the potential to be misused, especially if it
falls into the wrong hands.
Today the committee will hear from Federal witnesses on
this important topic. I am pleased that we have witnesses from
Customs and Border Protection, the Transportation Security
Administration, the Secret Service, and the National Institute
of Standards and Technology before us. They represent just a
few of the agencies involved in the Government's increasing use
of biometric technology.
I look forward to hearing from them about how they are
using biometric technology currently, their future plans, and
what they are doing to address these concerns. As Congress, it
is our job to ensure they protect the rights of the American
people before they move forward. I expect a good conversation
toward that end today and continued oversight by the committee
in the future.
[The statement of Chairman Thompson follows:]
Statement of Chairman Bennie G. Thompson
July 10, 2019
The Government's use of biometrics is not entirely new. For
example, fingerprints have been used as an identification tool for many
decades. Other biometrics include DNA, irises, voice patterns, and palm
prints. In recent years, facial recognition has become the new, chosen
form of biometric technology. As facial recognition technology has
advanced, its use by the Government and the private sector has also
increased. Currently, DHS is collecting and storing several different
kinds of biometric information and is using this information for
multiple purposes. CBP and TSA are using biometrics to confirm the
identities of travelers, for example. The Secret Service is piloting a
surveillance system using facial recognition. I am not opposed to
biometric technology, and recognize it can be valuable to homeland
security and facilitation. However, its proliferation across DHS raises
serious questions about privacy, data security, transparency, and
accuracy. The American people deserve answers to those questions before
the Federal Government rushes to deploy biometrics further.
Last month, the committee held roundtable discussions with both
industry and privacy and civil liberty stakeholders about the
Department of Homeland Security's increasing use of biometric
technology. Stakeholders have significant concerns about the data DHS
is collecting and whether the Department is safeguarding our rights
appropriately. They have good reason to be concerned. Absent standards,
Americans may not know when, where, or why the Department is collecting
their biometrics. People also may not know that they have the right to
opt out, or how to do so. Worse yet, they may not know that biometric
technology is in use, as is the case when face recognition is used to
passively surveil a crowd like under the Secret Service's pilot
program. Recent reports also indicate ICE has been scanning through
millions of Americans' drivers' license photos without their knowledge
or consent. These troubling reports are a stark reminder that biometric
technologies should only be used for authorized purposes in a fully
transparent manner.
Data security is another important concern. Frankly, the Federal
Government does not have a great track record securing Americans'
personal data, and biometric information can be particularly sensitive.
Unfortunately, earlier this year, a CBP subcontractor experienced a
significant data breach, including traveler images, raising important
questions about data security. Americans want to know that if the
Government collects their biometric data, they are going to keep it
secure from hackers and other bad actors. Moreover, the accuracy of
certain biometric technology is in question, despite advancement in
recent years. Studies by highly regarded academic institutions have
found facial recognition systems in particular are not as accurate for
women and darker-skinned individuals. Last July, the American Civil
Liberties Union (ACLU) conducted a test using Amazon's facial
recognition tool called ``Rekognition.'' The ACLU built a database of
25,000 publicly available arrest photos. Using Rekognition, the ACLU
searched the database using pictures of every current Member of
Congress. The software incorrectly matched 28 Members with individuals
who had criminal records. Although the misidentified members included
both Democrats and Republicans, men and women, and a wide range of
ages, nearly 40 percent of the false matches were people of color. This
is unacceptable.
It is not fair to expect certain people in our society to shoulder
a disproportionate burden of the technology's shortcomings. Before the
Government deploys these technologies further, they must be scrutinized
and the American public needs to be given a chance to weigh in.
Biometrics and facial recognition technology may be a useful homeland
security and facilitation tool, but as with any tool it has the
potential to be misused--especially if it falls into the wrong hands.
Today, the Committee will hear from Federal witnesses on this important
topic. I am pleased that we have witnesses from Customs and Border
Protection (CBP), the Transportation Security Administration (TSA), the
Secret Service, and the National Institute of Standards and Technology
(NIST) before us. They represent just a few of the agencies involved in
the Government's increasing use of biometric technology. I look forward
to hearing from them about how they are using biometric technology
currently, their plans for the future, and what they are doing to
address these concerns. As Congress, it is our job to ensure they
protect the rights of the American people before they move forward.
Chairman Thompson. I ask unanimous consent to enter the
following news articles and letters into the hearing's record:
A June 10 Washington Post article entitled ``U.S. Customs and
Border Protection Say Its Photos of Travelers Were Taken in a
Data Breach''; a July 7 Washington Post article entitled ``FBI
ICE Find State Driver's Licenses Photos Are a Gold Mine of
Facial Recognition Searches''; and July 9 letters from American
Association of Airport Executives, International Biometric
Identity Association, and a coalition of privacy and civil
liberties groups, many of whom were represented in our meetings
and briefings last month.
Without objection, so ordered.
[The information follows:]
U.S. Customs and Border Protection says photos of travelers were taken
in a data breach
By Drew Harwell and Geoffrey A. Fowler, June 10, 2019 at 7:54 p.m. EDT,
Washington Post.
U.S. Customs and Border Protection officials said Monday that
photos of travelers had been compromised as part of a ``malicious
cyberattack,'' raising concerns over how Federal officials' expanding
surveillance efforts could imperil Americans' privacy.
Customs officials said in a statement Monday that the images, which
included photos of people's faces and license plates, had been
compromised as part of an attack on a Federal subcontractor.
CBP makes extensive use of cameras and video recordings at airports
and land border crossings, where images of vehicles are captured. Those
images are used as part of a growing agency facial-recognition program
designed to track the identity of people entering and exiting the U.S.
Fewer than 100,000 people were impacted, said CBP, citing ``initial
reports.'' The photographs were taken of people in vehicles entering
and exiting the U.S. over a month and a half through a single land
border entry port, which CBP did not name. Officials said the stolen
information did not include other identifying information, and no
passport or other travel document photos were compromised.
The agency learned of the breach on May 31 and said that none of
the image data had been identified ``on the Dark Web or Internet.'' But
reporters at The Register, a British technology news site, reported
late last month that a large haul of breached data from the firm
Perceptics was being offered as a free download on the dark web.
CBP would not say which subcontractor was involved. But a Microsoft
Word document of CBP's public statement, sent Monday to Washington Post
reporters, included the name ``Perceptics'' in the title: ``CBP
Perceptics Public Statement.''
Perceptics representatives did not immediately respond to requests
for comment.
CBP spokeswoman Jackie Wren said she was ``unable to confirm'' if
Perceptics was the source of the breach.
One U.S. official, who spoke on condition of anonymity due to lack
of authorization to discuss the breach, said it was being described
inside CBP as a ``major incident.'' The official said Perceptics was
attempting to use the data to refine its algorithms to match license
plates with the faces of a car's occupants, which the official said was
outside of CBP's sanctioned use. The official said the data involved
travelers crossing the Canadian border.
The breach, according to the official, did not involve a foreign
nation, such as when China hacked the Office of Personnel Management in
2014 exposing the sensitive information of at least 22 million people.
News of the breach raised alarms in Congress, where lawmakers have
questioned whether the government's expanded surveillance measures
could threaten constitutional rights and open millions of innocent
people to identity theft.
``If the government collects sensitive information about Americans,
it is responsible for protecting it--and that's just as true if it
contracts with a private company,'' Sen. Ron Wyden (D-Ore.) said in a
statement to The Post. ``Anyone whose information was compromised
should be notified by Customs, and the government needs to explain
exactly how it intends to prevent this kind of breach from happening in
the future.''
Wyden said the theft of the data should alarm anyone who has
advocated expanded surveillance powers for the government. ``These vast
troves of Americans' personal information are a ripe target for
attackers,'' he said.
Civil rights and privacy advocates also called the theft of the
information a sign that the government's growing database of
identifying imagery had become an alluring target for hackers and
cybercriminals.
``This breach comes just as CBP seeks to expand its massive face
recognition apparatus and collection of sensitive information from
travelers, including license plate information and social media
identifiers,'' said Neema Singh Guliani, senior legislative counsel at
the American Civil Liberties Union. ``This incident further underscores
the need to put the brakes on these efforts and for Congress to
investigate the agency's data practices. The best way to avoid breaches
of sensitive personal data is not to collect and retain it in the first
place.''
CBP said copies of ``license plate images and traveler images
collected by CBP'' had been transferred to the subcontractor's company
network, violating the agency's security and privacy rules. The
subcontractor's network was then attacked and breached. No CBP systems
were compromised, the agency said.
Perceptics and other companies offer automated license-plate-
reading devices that Federal officials can use to track a vehicle, or
its owner, as it travels on public roads.
Immigration agents have used such databases to track down people
who may be in the country illegally. Police agencies have also used the
data to look for potential criminal suspects.
Perceptics, based in Tennessee, has championed its technology as a
key part of keeping the border secure. ``You want technology that
generates data you can trust and delivers it when and where you need it
most,'' a marketing website says.
The company also said recently that it had installed license-plate
readers at 43 U.S. Border Patrol checkpoint lanes across Arizona,
California, New Mexico and Texas, saying they offered border guards
``superior images with the highest license plate read rate accuracy in
North America.''
The Federal Government, as well as the group of private contractors
it works with, has access to a swelling database of people's cars and
faces, which it says is necessary to enhance security and enforce
border laws.
The FBI has access to more than 640 million photos, including from
passports and driver's licenses, that it can scan with facial-
recognition systems while conducting criminal investigations, a
representative for the Government Accountability Office told the House
Committee on Oversight and Reform at a hearing last week.
Rep. Bennie Thompson (D-Miss.), chair of the House Homeland
Security Committee, said he intended to hold hearings next month on
Homeland Security's use of biometric information.
``Government use of biometric and personal identifiable information
can be valuable tools only if utilized properly. Unfortunately, this is
the second major privacy breach at DHS this year,'' Thompson said,
referring to a separate breach in which more than 2 million U.S.
disaster survivors had their information revealed by the Federal
Emergency Management Agency. ``We must ensure we are not expanding the
use of biometrics at the expense of the privacy of the American public.
``
Nick Miroff, Ellen Nakashima and Tony Romm contributed to this
report.
______
FBI, ICE find State driver's license photos are a gold mine for facial-
recognition searches
By Drew Harwell, July 7, 2019 at 3:54 p.m. EDT, The Washington Post
Agents with the Federal Bureau of Investigation and Immigration and
Customs Enforcement have turned State driver's license databases into a
facial-recognition gold mine, scanning through millions of Americans'
photos without their knowledge or consent, newly released documents
show.
Thousands of facial-recognition requests, internal documents and
emails over the past 5 years, obtained through public-records requests
by researchers with Georgetown Law's Center on Privacy and Technology
and provided to The Washington Post, reveal that Federal investigators
have turned state departments of motor vehicles data bases into the
bedrock of an unprecedented surveillance infrastructure.
Police have long had access to fingerprints, DNA and other
``biometric data'' taken from criminal suspects. But the DMV records
contain the photos of a vast majority of a state's residents, most of
whom have never been charged with a crime.
Neither Congress nor state legislatures have authorized the
development of such a system, and growing numbers of Democratic and
Republican lawmakers are criticizing the technology as a dangerous,
pervasive and error-prone surveillance tool.
``Law enforcement's access of state databases,'' particularly DMV
databases, is ``often done in the shadows with no consent,'' House
Oversight Committee Chairman Elijah E. Cummings (D-Md.) said in a
statement to The Post.
Rep. Jim Jordan (Ohio), the House Oversight Committee's ranking
Republican, seemed particularly incensed during a hearing into the
technology last month at the use of driver's license photos in Federal
facial-recognition searches without the approval of state legislators
or individual license holders.
``They've just given access to that to the FBI,'' he said. ``No
individual signed off on that when they renewed their driver's license,
got their driver's licenses. They didn't sign any waiver saying, `Oh,
it's OK to turn my information, my photo, over to the FBI.' No elected
officials voted for that to happen.''
Despite those doubts, Federal investigators have turned facial
recognition into a routine investigative tool. Since 2011, the FBI has
logged more than 390,000 facial-recognition searches of Federal and
local databases, including state DMV databases, the Government
Accountability Office said last month, and the records show that
Federal investigators have forged daily working relationships with DMV
officials. In Utah, FBI and ICE agents logged more than 1,000 facial-
recognition searches between 2015 and 2017, the records show. Names and
other details are hidden, though dozens of the searches are marked as
having returned a ``possible match.''
San Francisco and Somerville, Mass., have banned their police and
public agencies from using facial-recognition software, citing concerns
about governmental overreach and a breach of public trust, and the
subject is being hotly debated in Washington. On Wednesday, officials
with the Transportation Security Administration, Customs and Border
Protection and the Secret Service are expected to testify at a hearing
of the House Committee on Homeland Security about their agencies' use
of the technology.
The records show the technology already is tightly woven into the
fabric of modern law enforcement. They detailed the regular use of
facial recognition to track down suspects in low-level crimes,
including cashing a stolen check and petty theft. And searches are
often executed with nothing more formal than an email from a Federal
agent to a local contact, the records show.
``It's really a surveillance-first, ask-permission-later system,''
said Jake Laperruque, a senior counsel at the watchdog group Project on
Government Oversight. ``People think this is something coming way off
in the future, but these [facial-recognition] searches are happening
very frequently today. The FBI alone does 4,000 searches every month,
and a lot of them go through state DMVs.''
The records also underscore the conflicts between the laws of some
states and the Federal push to find and deport undocumented immigrants.
Though Utah, Vermont and Washington allow undocumented immigrants to
obtain full driver's licenses or more-limited permits known as driving
privilege cards, ICE agents have run facial-recognition searches on
those DMV databases.
More than a dozen states, including New York, as well as the
District of Columbia, allow undocumented immigrants to drive legally
with full licenses or driving privilege cards, as long as they submit
proof of in-state residency and pass the states' driving-proficiency
tests.
Lawmakers in Florida, Texas and other states have introduced bills
this year that would extend driving privileges to undocumented
immigrants. Some of those states already allow the FBI to scan driver's
license photos, while others, such as Florida and New York, are
negotiating with the FBI over access, according to the GAO.
``The state has told [undocumented immigrants], has encouraged
them, to submit that information. To me, it's an insane breach of trust
to then turn around and allow ICE access to that,'' said Clare Garvie,
a senior associate with the Georgetown Law center who led the research.
An ICE spokesman declined to answer questions about how the agency
uses facial-recognition searches, saying its ``investigative techniques
are generally considered law-enforcement sensitive.''
Asked to comment, the FBI referred The Post to the congressional
testimony last month of Deputy Assistant Director Kimberly Del Greco,
who said that facial-recognition technology was critical ``to preserve
our nation's freedoms, ensure our liberties are protected, and preserve
our security.'' The agency has said in the past that while facial-
recognition searches can provide helpful leads, agents are expected to
verify the findings and secure definitive proof before pursuing arrests
or criminal charges.
Twenty-one states, including Texas and Pennsylvania, plus the
District of Columbia, allow Federal agencies such as the FBI to scan
driver's license photos, GAO records show. The agreements stipulate
some rules for the searches, including that each must be relevant to a
criminal investigation.
The FBI's facial-recognition search has access to local, state and
Federal databases containing more than 641 million face photos, a GAO
director said last month. But the agency provides little information
about when the searches are used, who is targeted and how often
searches return false matches.
The FBI said its system is 86 percent accurate at finding the right
person if a search is able to generate a list of 50 possible matches,
according to the GAO. But the FBI has not tested its system's accuracy
under conditions that are closer to normal, such as when a facial
search returns only a few possible matches.
Civil rights advocates have said the inaccuracies of facial
recognition pose a heightened danger of misidentification and false
arrests. The software's precision is highly dependent on a number of
factors, including the lighting of a subject's face and the quality of
the image, and research has shown that the technology performs less
accurately on people with darker skin.
``The public doesn't have a way of controlling what information the
government has on them,'' said Jacinta Gonzalez, a senior organizer for
the advocacy group Mijente who was particularly concerned about how ICE
and other agencies could use the scans to track down immigrants. ``And
now there's this rapidly advancing technology, with very few guidelines
and protections for people, putting all of this information at their
fingertips in a very scary way.''
The records, which include thousands of emails and official
documents from Federal agencies, as well as Utah, Vermont and
Washington State, show how easy it is for a Federal investigator to tap
into an individual State DMV's database. While some of the driver photo
searches were made on the strength of Federal subpoenas or court
orders, many requests for searches involved nothing more than an email
to a DMV official with the target's ``probe photo'' attached. The
official would then search the driver's license database and provide
details of any possible matches.
The search capability was offered not just to help identify
criminal suspects, but also to detect possible witnesses, victims,
bodies, and innocent bystanders and other people not charged with
crimes.
Utah's DMV database was the subject of nearly 2,000 facial-
recognition searches from outside law enforcement agencies between 2015
and 2017--sometimes dozens of searches a day, the records show. One
document from Utah's Statewide Information & Analysis Center coached
officers on how to make facial-recognition requests; offered four tips
for better facial photographs (``lighting, distance, angle, eyes'');
and said the database included ``over 5 million Utah driver's license &
state identification card photos,'' about 2 million more than the
state's population. State officials did not respond to requests for
comment.
Many of the requests for searches in Utah came from local police
forces across the country seeking to find suspects who may have
traveled to the state, but roughly half the searches came from Federal
agents, according to a log of the searches. The records do not provide
suspect names or say whether cases ended in arrests or convictions.
Washington state's Department of Licensing said that its ``facial
recognition system is designed to be an accurate, non-obtrusive fraud
detection tool'' and that the agency does not share use of the system
with law enforcement unless compelled by a court order.
Vermont officials said they stopped using facial-recognition
software in 2017. That year, a local chapter of the American Civil
Liberties Union revealed records showing that the state DMV had been
conducting the searches in violation of a state law that banned
technology involving ``the use of biometric identifiers.'' The state's
Governor and attorney general came out against the face-scanning
software, citing a need to balance public safety with residents'
privacy rights.
In the years before the ban, the records show, Vermont officials
ran a number of face scans on driver's license photos at the request of
ICE agents. Investigators from a number of Federal and local agencies
emailed the state's DMV with facial-recognition search requests as they
pursued people accused of overstaying their visas, providing false
information, stealing from stores or, in at least one case, being part
of a ``suspicious circumstance.''
The officers in some emails would provide descriptions of their
targets: One was dubbed a ``gypsy . . . scamming elderly people for
money,'' while another was said to have ``VERY LARGE PROTRUDING EARS.''
In others, DMV officials talked about the face-scanning tool as if it
were the kind of awe-inspiring technical marvel most often seen on
prime-time cop shows.
In one 2014 email, a police officer in the town of Manchester, Vt.,
asked a DMV official to scan for a man caught on video ``brazenly''
stealing. The official forwarded the email to a colleague with a made-
for-TV flourish, writing, ``Can we play NCIS for this officer?''
______
Letter From Todd Hauptli, American Association of Airport Executives
July 9, 2019.
The Honorable Bennie Thompson,
Chairman, House Homeland Security Committee, Washington, DC 20515.
The Honorable Mike Rogers,
Ranking Member, House Homeland Security Committee, Washington, DC
20515.
Dear Chairman Thompson and Ranking Member Rogers: As experiences at
airports across the country and the world illustrate, biometric
technology holds tremendous promise in enhancing security and
efficiency in the aviation environment. While airport executives are
encouraged by the promise of biometrics and look forward to their
further utilization, we recognize that there are legitimate privacy and
civil right concerns that must be addressed before these technologies
are deployed more widely. We look forward to working with the Committee
and the Department of Homeland Security (DHS) to ensure that the proper
regulatory framework and safeguards are in place to protect citizens'
rights as these technologies are utilized to achieve worthy objectives.
As you know, many airport facilities across the country are already
experiencing significant strain with passenger traffic at record
levels. The situation is likely to become increasingly more challenging
as airport facilities age and as domestic and international passenger
levels continue to increase. International passenger traffic is growing
at 5-6 percent at U.S. airports and domestic growth is nearing 5
percent, with some facilities seeing growth well beyond those annual
averages. International air travel is projected to double over the next
20 years according to the International Air Transportation Association
(IATA).
While there are clear economic benefits that accompany these
increases in passengers, airport facilities--many of which are decades
old--cannot keep pace with current growth. Similarly, our Federal
partners at both U.S. Customs and Border Protection and the
Transportation Security Administration will, undoubtedly, have
significant difficulties handling record passenger volumes efficiently
and effectively at current staffing levels, leading to growing wait
times at checkpoints and in other processing queues.
In our view, innovation holds the key to improving the efficiency
of the travelers' journey and reducing growing lines, which themselves
pose a security challenge. Wider adoption of biometric technology at
our borders and security checkpoints is one way that airports,
airlines, and the Federal Government can more seamlessly handle
expected passenger growth. Biometrics, including facial recognition,
have the potential to enhance security and efficiency without
compromising important civil liberties provided that their utilization
is coupled with robust privacy and data protections for travelers and
the ability for American citizens to opt out of using biometric
technology in favor of the traditional screening process at an airport.
Additionally, as Federal budgets are tightening, we are concerned
that DHS may shift the responsibility for acquiring these technologies
onto airports at a time when State and local budgets are also
constricting. This could lead to a bifurcated system in which certain
airports or airlines have the financial resources to procure these
biometric technologies and others do not, resulting in different
protocols being used at different airports in the United States. We are
already seeing the Department depend heavily on public-private
partnerships to fund inherently governmental responsibilities for the
screening and processing of passengers. Security cannot and should not
become an area of ``haves'' and ``have nots'' at the nation's airports.
AAAE and our members would welcome the opportunity to discuss
potential uses and security benefits of biometric and facial
recognition technologies in the airport environment as you contemplate
further Committee action in this area. We sincerely appreciate your
consideration of our views and the need to innovate in order to address
growing passenger volumes at our airports while maintaining the highest
levels of security.
Sincerely,
Todd Hauptli,
President and CEO.
______
Statement of the International Biometrics + Identity Association
July 9, 2019.
The Honorable Bennie Thompson,
Chairman, Committee on Homeland Security, U.S. House of
Representatives, 2466 Rayburn House Office Building,
Washington, DC 20515.
The Honorable Mike Rogers,
Ranking Member, Committee on Homeland Security, U.S. House of
Representatives, 2184 Rayburn House Office Building,
Washington, DC 20515.
Dear Chairman Thompson and Ranking Member Rogers: On behalf of the
International Biometrics + Identity Association, I am writing to
express our gratitude for your support over the past decade for the use
of biometrics by DHS, and especially CBP for its US-VISIT Entry Exit
program. Through your efforts, CBP has come a long way in implementing
biometrics to enhance the security of air travel, limit identification
fraud, help address the visa overstay issue, and, at the same time,
facilitate air travel.
A heated debate is now surrounding the emergence of facial
recognition for use by CBP, TSA, and other DHS programs.
IBIA appreciates the opportunity to provide the following
information about the performance of facial recognition algorithms that
has not come out in the public hearings. The data that we are providing
comes directly from NIST, the recognized global premier testing entity.
IBIA acknowledges that many people have concerns about privacy that
are rooted in moral and political philosophies. These are matters of
opinion on which reasonable people may disagree and should be resolved
in the public sphere.
IBIA's objective is to provide facts that can help to inform the
debate and conversations about facial recognition, facts that have not
been properly aired to date. This is foundational for good legislation.
We look to this Committee to help bring out the facts to ensure a full
debate on the issues open to all stakeholders and relevant information.
Much of the debate has centered around the view that the algorithms
are ``biased'' and ``discriminatory''. These words are semantically
loaded and imply intent. Facial recognition is performed by a machine
and machines have no intent.
Attachment.--International Biometrics + Identity Association
understanding the performance of facial recognition algorithms
The International Biometrics + Identity Association (IBIA) is the
leading voice for the biometrics and identity technology industry. It
advances the transparent and secure use of these technologies to
confirm human identity in our physical and digital worlds.
#identitymatters
understanding the performance of facial recognition algorithms
Executive Summary
This paper addresses the performance of facial recognition
algorithms, an issue that has emerged as a major point of contention
during the current policy debates about the use and limits of facial
recognition.
The thrust of the argument to limit the use of facial recognition
is that the technology is not yet ready for prime time. The primary
arguments are that facial recognition algorithms are basically too
imperfect because they are ``discriminatory'' against people with dark
skin tones and display low levels of matching performance.
The purposes of this paper are to:
Demonstrate these performance arguments are not supported by
the evidence documented in recent National Institute of
Standards and Technology (NIST) testing, the world's premier
standards and testing body. NIST shows stunningly high levels
of accuracy and clear superiority of the technology compared to
human recognition systems, both in terms of accuracy rates and
performance across a range of skin tones. This is supported by
the latest academic research conducted by a group of the
preeminent scholars on facial recognition.
Explain the factors that affect the performance differences
of facial recognition algorithms, including the application,
the rest of the system, variations in quality of the
algorithms.
Summarize the many benefits of facial recognition.
Highlight the challenges in the use of facial recognition
that remain and address the work in progress to further improve
the technology.
The field of research today known as Artificial Intelligence traces
its origins to a workshop at Dartmouth College in 1956. Attendees
became the founders and leaders of the field and were, with the benefit
of hindsight, unrealistic about the likely course of progress. For
example, Herbert Simon predicted, ``machines will be capable, within 20
years, of doing any work a man can do.'' Marvin Minsky agreed, writing
``within a generation . . . the problem of creating `artificial
intelligence' will substantially be solved.'' What AI research has
delivered are highly specialized tools which approximate or improve
upon human performance in narrow areas, yet exhibit no generalized
behavior that humans would recognize as intelligence. Deep Learning is
another such term that implies an on-going process similar to that
employed by humans; whereas what actually occurs is a highly
sophisticated, one time, training on substantial amounts of carefully
annotated data. Thereafter the system works well with information
similar to the training data but does not adapt to new data until a
subsequent training period.
Let's Stop Using Semantically Loaded Terms like ``Discriminatory''
Let's dispense with this term so we can focus on the
essential facts about performance of facial recognition
systems, including accuracy and systemic errors, instead of
extraneous and emotional issues.
``Discriminatory'' is a semantically loaded term because it
implies intent.
However, facial recognition is performed by a machine, and
machines have no intent.
The argument that algorithm developers exhibit racial/gender
blindness producing algorithms that perform less effectively
for other than white males is not supported by the facts.
NIST has active test and analysis effort to assess this
issue.
Recent (12 April 2019) results for verification algorithms
(i.e. 1:1 search) show the top 20 performing algorithms, with
elapsed time between images from 2-16 years, are most effective
for blacks with black females often the most accurate.
The test results for identification (i.e. 1:N search) are
expected during 2Q 2019.
The most appropriate composition of test datasets, to insure
effective testing, is still somewhat of an unsettled issue.
Cost of new dataset development for effective large-scale testing
is a significant issue, beyond the resources of all but government and
the largest companies. It may be feasible to continue to employ
existing facial recognition datasets, by recharacterizing their
metadata to more accurately reflect subject demographics, once there is
consensus on what changes, if any, are needed.
Performance Differences of Algorithms
All algorithms have some performance differences across
different demographic groups, genders, and age cohorts.
These differences are being addressed and there has been
rapid improvement, which is on-going.
For verification applications (fraud detection, access
control, etc.), in the latest NIST testing, the top performing
algorithms are more accurate with black males and females than
with whites and have less than 1 percent false non-match rates
for all groups at 0.1 percent false match rate.
For investigative applications, progress has been dramatic
with a major update report expected from NIST during the 2d
quarter of 2019.
Facial Recognition and Facial Classification are Different and Should
Not Be Confused
Face recognition seeks to identify an individual from their
face image.
Facial classification seeks to classify a face by
estimating, for example, gender, age, or race.
The algorithms are built and trained separately.
The process of classification estimation involves one image,
while facial recognition involves comparison of pairs.
An MIT study, which is a large part of the ``facial
recognition is biased narrative''; only examined facial
classification, specifically for gender.
A joint FIT/Notre Dame study provides a more complete and
accurate view, as do the NIST tests.
Algorithms Are Just Part of a Facial Recognition System
The performance of a facial recognition system depends on a
number of factors; the algorithm is one such factor. The
camera, its resolution, positioning, distance, and lighting set
an upper limit on performance. Subject pose and expression can
also influence performance.
Camera resolution and distance matter; humans require about
25 pixels per meter resolution to detect the presence of
humans, but can recognize motion at lower resolutions.
Ambient or artificial lighting has an enormous impact on
system performance.
In other words, all the components of the facial recognition
system must perform properly, in addition to using a high-
performance algorithm, and these elements can be adjusted
easily.
Knowing all this, some facial recognition applications
employ human facial examiners who make the final match/no match
decision after the facial matching algorithm selects a list of
potential matches; they use applications specifically designed
for facial examinations.
The Application Matters
Facial verification and facial identification systems, until
quite recently, have been designed to match portrait style
(mugshot, driver license, visa, passport) images.
With good lighting, pose, and expression control,
performance can be stunningly good and good mugshot accuracy
conforms to photography standards adopted by NIST for the FBI
further developed by ISO.
Matching of ``in the wild'' images (a reference to image
quality--candid, unposed, not portrait-style images) has
matured dramatically in the past 5 years, with verification
accuracy of top algorithms now at 99 percent. An update on
investigation applications is expected to show comparable
progress and further maturation is expected in the near term.
Some algorithms are much better than others, as in everything else.
In golf, there is Tiger Woods and then there is the rest of us.
Not all Algorithms are Alike
Market entry is relatively easy and the number of algorithm
providers has expanded from about 10 in 2010 to about 100
today, with many offering multiple algorithms.
Some algorithms are much better than others, as would be
expected. Objective testing like that performed by NIST reveals
the differences.
Algorithm performance for a selfie, social media, or a
commodity web camera is considerably different from an
algorithm used for security or law enforcement applications.
NIST Has Tested More Than 170 FR Algorithms, with Wide Variations in
Performance Observed
Six (6) algorithms are less accurate than a coin toss.
Most are more accurate than human observers, including those
trained and employed to do recognition.
The top performing algorithms are much better performing
than humans.
Many algorithms match blacks more accurately than whites.
Algorithm matching of females is frequently less accurate
than males.
Algorithm performance is less accurate for most applications
involving children.
The application makes a difference.
Portrait style 1:N and 1:1 matching is extraordinarily
accurate (considerably more accurate than fingerprint
technology circa mid-2000's when FBI went to partial ``lights
out'' fingerprint matching).
Nothing is perfect and no system performs perfectly. The real
question is whether automated facial recognition is better than the
current systems. And under this criterion, data clearly demonstrates
superior performance of automated facial recognition.
In the wild (``candid, unposed, non-portrait images''),
matching is less accurate but quite suitable for lead
generation, typically with stalled investigations.
Likewise, matching is less accurate for poor quality images.
Notwithstanding exceptional algorithm accuracy, validation
has not been performed to allow ``lights out'' use of facial
recognition technology when there are potential adverse
consequences to the subjects. Human review is required.
Algorithms are not commoditized as performance varies
greatly, from the best identifying 99.4 percent of individuals
in a gallery of 12 million subjects to below 40 percent for the
worst.
Demand for Perfection of Algorithms is Not a Performance Standard for
the Real World
No system--or human--performs perfectly.
The real question is whether automated facial recognition is
better than other systems or humans. And under this criterion,
data clearly demonstrate superior performance of automated
facial recognition.
For family, friends, professional acquaintances, and
celebrities, human recognition works well.
For unfamiliar persons, few individuals perform well at face
recognition or matching.
Skilled passport examiners are only about 80 percent
accurate when unaided by automation.
The top performing algorithms outperform the mean
performance of all human groups including skilled forensic face
examiners with unlimited time and the best automated tools;
(although a few humans in the more skilled groups outperform
circa 2017 top algorithms).
Machines can memorize millions of faces, and humans only
thousands, enabling machines to do things unaided that humans
cannot, including to:
Identify missing children who do not know their names.
Identify exploited children in dark web pornography.
Identify disoriented adults (e.g. with amnesia,
Alzheimer's).
Flag likely driver license application fraud for human
review.
Identify likely Visa fraud for human review.
Identify likely Passport fraud for human review.
Provide leads for further investigation when a
surveillance photo is the only information.
Detect border (and other) fraudulent use of stolen
identity documents.
People are Comfortable with Face Recognition
Following the iPhone X introduction on November 3, 2017,
tens of millions of Americans have become familiar and entirely
satisfied with facial recognition technology for personal use.
The 2019 Center for Data Innovation public opinion survey
found that only 1 in 4 Americans think the government should
strictly limit the use of facial recognition technology.
The technology is widely used worldwide, and adoption is
growing.
DHS pilot projects at several airports, dispensing with
boarding passes and ID cards in favor of facial recognition for
international flights, have been enthusiastically greeted by
the traveling public.
Frequent international travelers already hope for domestic
adoption.
Technology advancement is inexorable, and each generation
has the responsibility to decide how to balance the benefits of
new technology with privacy and appropriate uses.
The IBIA is the leading voice for the biometrics and identity
technology industry. It advances the transparent and secure use of
these technologies to confirm human identity in our physical and
digital worlds. Visit us at www.ibia.org.
______
Letter Submitted By Chairman Bennie G. Thompson
July 9, 2019.
The Honorable Bennie Thompson,
Chairman, Committee on Homeland Security, 310 Cannon House Office
Building, Washington, DC 20515.
The Honorable Mike Rogers,
Ranking Member, Committee on Homeland Security, 310 Cannon House Office
Building, Washington, DC 20515.
RE: The Suspension of Face Recognition Technology Use by the Department
of Homeland Security
Dear Chairman Thompson and Ranking Member Rogers: The undersigned
organizations, which are dedicated to preserving privacy, civil
liberties, and civil rights, write to urge you to immediately suspend
the Department of Homeland Security's (DHS) use of face recognition
technology on the general public.
The use of face recognition technology by the DHS poses serious
risks to privacy and civil liberties, threatens immigrants, broadly
impacts American citizens, and has been implemented without proper
safeguards in place or explicit congressional approval. The technology
is being deployed today by authoritarian governments as a tool to
suppress speech and monitor critics, minorities, and everyday citizens.
Congress should not permit the continued use of face recognition in the
United States absent safeguards to prevent such abuses.
Moreover, the extraordinary breach of the images of travelers'
faces and license plates, surveillance-equipment schematics and
sensitive contracting documents by a CBP contractor has made clear that
these programs are creating new risks to the privacy and security of
Americans.\1\ Through carelessly managed programs, DHS itself created
new security threats. It would be irresponsible for DHS to move forward
with face recognition programs that collect massive amounts of
sensitive data until a thorough investigation of this incident is
completed and the agency demonstrates that it can fully safeguard its
systems.
---------------------------------------------------------------------------
\1\ Drew Harwell, Hacked documents reveal sensitive details of
expanding border surveillance, Wash. Post (June 21), https://
www.washingtonpost.com/technology/2019/06/21/hacked-documents-reveal-
sensitive-details-expanding-border-surveillance/.
---------------------------------------------------------------------------
DHS's Use of Face Recognition Technology
DHS is in the process of integrating and expanding the agency's use
of face recognition technology through various programs of its
subcomponents. DHS's use of face recognition will affect millions of
individuals, who will lack the protections needed against a powerfully
invasive surveillance tool.
Customs and Border Protection
The broadest current use of face recognition technology is the
Customs and Border Protection's Biometric Entry-Exit program. Without
legal authority or the opportunity for public comment, the U.S. Customs
and Border Protection (CBP) has broadly deployed facial recognition
technology at U.S. airports to all travelers, including U.S. citizens.
The agency plans to ``incrementally deploy biometric capabilities
across all modes of travel--air, sea, and land--by fiscal year
2025.''\2\
---------------------------------------------------------------------------
\2\ U.S. Dep't of Homeland Sec., Office of Inspector Gen., OIG-18-
80, Progress Made, but CBP Faces Challenges Implementing a Biometric
Capability to Track Air Passenger Departures Nationwide, 7 (Sept. 21,
2018), https://www.oig.dhs.gov/sites/default/files/assets/2018-09/OIG-
18-80-Sep18.pdf [hereinafter OIG Report].
---------------------------------------------------------------------------
CBP uses flight manifests and photographs obtained from the State
Department to create ``galleries'' to match with photos captured at
international airports.\3\ ``If CBP does not have access to advance
passenger information, such as for pedestrians or privately-owned
vehicles at land ports of entry, CBP will build galleries using
photographs of `frequent' crossers for that specific port of
entry[.]''\4\ CBP uses its own equipment as well as that of private
firms, other government agencies, and foreign governments to capture
face images.\5\ Yet, there are no formal rules restricting the use of
the photos captured by non-CBP owned equipment.\6\
---------------------------------------------------------------------------
\3\ Id.
\4\ U.S. Dep't of Homeland Sec., U.S. Customs and Border
Protection, DHS/CBP/PIA-0056, Privacy Impact Assessment for the
Traveler Verification Service, 5 (Nov. 14, 2018) https://www.dhs.gov/
sites/default/files/publications/privacy-pia-cbp030-tvs-
november2018_2.pdf [hereinafter TVS Nov. 2018 PIA].
\5\ Id. at 7-8.
\6\ See Memorandum of Understanding Between and Among U.S. Customs
and Border Protection and [REDACTED] and [REDACTED] Regarding
[REDACTED] Biometric Pilot Project at [REDACTED] (June 2017), https://
epic.org/foia/dhs/cbp/biometric-entry-exit/MOU-Biometric-Pilot-
Project.pdf.
---------------------------------------------------------------------------
The steady implementation of CBP's biometric entry-exit program in
airports across the country has been widely reported.\7\ The program
affects a significantly large group of U.S. citizens traveling in and
out of the country. At the Atlanta Hartsfield-Jackson International
Airport alone, ``[a]bout 25,000 passengers move through the terminal
each week'' and the majority of those passengers are subject to facial
recognition.\8\ Further, ``CBP hopes to have facial recognition
boarding at all US airports serving international flights within 3 or 4
years.''\9\
---------------------------------------------------------------------------
\7\ See e.g., Bart Jansen, CBP: Orlando is First U.S. Airport to
Scan Faces of All International Travelers, USA Today (June 21, 2018),
https://www.usatoday.com/story/travel/flights/todayinthesky/2018/06/2/
orlando-international-airport-scan-faces-u-s-citizens/722643002/; Lori
Aratani, Officials Unveil New Facial Recognition System at Dulles
International Airport, Wash. Post (Sept. 7, 2018), https://
www.washingtonpost.com/transportation/2018/09/06/officials-unveil-new-
facial-recognition-system-dulles-international-airport/; Gregory
Wallace, Instead of the Boarding Pass, Bring Your Smile to the Airport,
CNN (Sept. 10, 2018), https://www.cnn.com/travel/article/cbp-facial-
recognition/index.html; Jack Stewart, Creepy or Not, Face Scans Are
Speeding Up Airport Security, Wired (Nov. 21, 2018), https://
www.wired.com/story/airport-security-biometrics-face-scanning/;.
\8\ Lori Aratani, Your Face is Your Boarding Pass at this Airport,
Wash. Post (Dec. 4, 2018), https://www.washingtonpost.com/nation/2018/
12/04/your-face-is-your-boarding-pass-this-airport/.
\9\ Thom Patterson, US Airport Opens First Fully Biometric
Terminal, CNN (Dec. 3, 2018), https://www.cnn.com/travel/article/
atlanta-airport-first-us-biometric-terminal-facial-recognition/
index.html.
---------------------------------------------------------------------------
The Biometric Entry-Exit program is flawed. A report on iris and
facial recognition technologies at a southern land border found that
the technologies did not perform operational matching at a
``satisfactory'' level.\10\ A DHS Office of the Inspector General
(``IG'') report found that CBP's Biometric Entry-Exit program suffered
from technical and operational challenges. The IG report also found
that CBP could not ``produce biometric matches consistently for
individuals in certain passenger groups'' with the lowest biometric
confirmation rate being for U.S. citizens.\11\ Moreover, several
reports and studies have noted that face recognition algorithms are
often less accurate on certain sub-groups, including women and people
with darker skin pigmentation.\12\
---------------------------------------------------------------------------
\10\ U.S. Customs and Border Protection, Southern Border Pedestrian
Field Test Summary Report, 8 (Dec. 2016), https://epic.org/foia/dhs/
cbp/biometric-entry-exit/Southern-Border-Pedestrian-Field-Test-
Report.pdf.
\11\ OIG Report at 19.
\12\ See, e.g., Joy Buolamwini & Timnit Gebru, Gender Shades:
Intersectional Accuracy Disparities in Commercial Gender
Classification, Proceedings of the 1st Conference on Fairness,
Accountability and Transparency, PMLR 81:77-91 (2018), http://
proceedings.mlr.press/v8l/buolamwini18a/buolamwini18a.pdf.
---------------------------------------------------------------------------
Americans returning to the United States have also found it
difficult to opt out of the facial recognition screening, which is
their legal right.\13\ Travelers routinely report on burdensome
procedures intended to compel individuals to undergo facial recognition
even if that is not their choice.\14\ Additionally, CBP has not
undergone formal rulemaking addressing how information collected will
be used, disclosed, and retained, and what remedies will exist in cases
where individuals are adversely impacted by the use of the technology.
---------------------------------------------------------------------------
\13\ See Zack Whittaker, Yes, Americans can opt-out of airport
facial recognition--here's how, Tech Crunch, https://techcrunch.com/
2019/05/13/americans-opt-out-facial-recognition-airport/; Allie Funk, I
Opted Out of Facial Recognition at the Airport--It Wasn't Easy, Wired,
July 2, 2019, https ://www.wired.com/story/opt-out-of-facial-
recognition-at-the-airport/.
\14\ Allie Funk, I Opted Out of Facial Recognition at the Airport--
It Wasn't Easy, Wired, July 2, 2019, https://www.wired.com/story/opt-
out-of-facial-recognition-at-the-airport/.
---------------------------------------------------------------------------
These concerns are further amplified given that CBP uses face
recognition technology for purposes that extend far beyond simply
verifying whether someone purportedly matches the photograph on their
travel document. CBP plans to use the facial recognition to search
biometric watch lists--raising questions about how such lists will be
compiled and whether they will be the predicate for additional
immigration and law enforcement activities \15\ The data from the
Biometric Entry-Exit program will also be broadly accessible within DHS
with the Coast Guard, Transportation Security Administration (TSA), and
Immigration and Customs Enforcement (ICE) all having access to the
data.\16\
---------------------------------------------------------------------------
\15\ U.S. Department of Homeland Security U.S. Customs and Border
Protection, Biometric Entry-Exit Program Concept of Operations, 000039
(June 27, 2017), https://epic.org/foia/dhs/cbp/biometric-entry-exit/
CBP-Biometric-Entry-Exit-Concept-of-Operations.pdf.
\16\ See U.S. Department of Homeland Security U.S. Customs and
Border Protection, Biometric Entry-Exit Program Concept of Operations
000063 (June 27, 2017), https://epic.org/foia/dhs/cbp/biometric-entry-
exit/Concept-of-Operations.pdf; see also U.S. Department of Homeland
Security, Capability Analysis Study Plan for Biometric Ently Exit
000160-000161, https://epic.org/foia/dhs/cbp/biometric-entry-exit/
Capability-Analysis-Study-Plan.pdf.
---------------------------------------------------------------------------
Transportation Security Administration
The TSA has plans to expand the use of face recognition to all
domestic travelers.\17\ The TSA Biometric Roadmap envisions the use of
face recognition for booking, check-in, bag drop, the security line,
access to an airport lounge, and boarding.\18\ TSA states it ``will
pursue a system architecture that promotes data sharing to maximize
biometric adoption throughout the passenger base and across the
aviation security touchpoints of the passenger experience.''\19\
---------------------------------------------------------------------------
\17\ Transportation Security Administration, TSA Biometrics Roadmap
(Sept. 2018), https://www.tsa.gov/sites/default/files/
tsa_biometrics_roadmap.pdf.
\18\ Id. at 18.
\19\ Id. at 17.
---------------------------------------------------------------------------
Similar to CBP, ISA has not undergone rulemaking clarifying how
information will be collected, used, or retained. However, TSA's
biometric roadmap suggests that its system will be interoperable with
CBP, and thus may be utilized for other immigration and law enforcement
activities.
Immigration and Customs Enforcement
Recent news reports show that ICE has expanded the agency's
deployment and use of face recognition systems. Public records covered
by the press this week show that ICE has been sending facial
recognition requests to State DMVs for years.\20\ As a result, millions
of innocent State residents have had their faces scanned by ICE without
notice or consent. Internal documents also suggest that ICE plans to
leverage CBP's biometric entry-exit system to identify and search for
information regarding non-citizens encountered during enforcement
activities.\21\
---------------------------------------------------------------------------
\20\ Drew Harwell, FBI, ICE find State driver's license photos are
a gold mine for facial-recognition searches, Wash. Post (July 7, 2019),
https://www.washingtonpost.com/technology/2019/07/07/fbi-ice-find-
state-drivers-license-photos-are-gold-mine-facial-recognition-
searches/.
\21\ See U.S. Department of Homeland Security U.S. Customs and
Border Protection, Biometric Entry-Exit Program: Concept of Operations,
000063 (June 2017), https://epic.org/foia/dhs/cbp/biometric-entry-exit/
CBP-Biometric-Entry-Exit-Concept-of-Operations.pdf.
---------------------------------------------------------------------------
In addition, last year, Amazon marketed the company's facial
recognition service ``Rekognition'' to ICE for border control.\22\ A
test of Amazon's face recognition software resulted in Amazon's
technology falsely matching 28 Members of Congress to mugshots and
other tests have similarly found the technology to be less accurate on
individuals with darker skin pigmentations.\23\
---------------------------------------------------------------------------
\22\ Drew Harwell, Amazon met with ICE officials over facial-
recognition system that could identify immigrants, Wash. Post (Oct. 23,
2018), https://www.washingtonpost.com/technology/2018/10/23/amazon-met-
with-ice-officials-over-facial-recognition-system-that-could-identify-
immigrants/.
\23\ Natasha Singer, Amazon's Facial Recognition Wrongly Identifies
28 Lawmakers, A.C.L.U. Says, N.Y. Times (July 26, 2018), https://
www.nytimes.com/2018/07/26/technology/amazon-aclu-facial-recognition-
Congress.html.
---------------------------------------------------------------------------
There is a lack of public information on how ICE might use the face
recognition capabilities implemented as part of the Biometric Entry-
Exit program, ICE's current use of face recognition technology, and
whether the agency intends to deploy other face recognition
capabilities. There is a serious risk that ICE could deploy face
recognition for purposes of indiscriminate immigration enforcement and
use the technology, despite its record of error, as a pretext for
aggressive questioning and harassment of immigrants--including those
lawfully present in the United States.
Secret Service
The U.S. Secret Service is testing the use of face recognition
technology to identify people in the public spaces in and around the
White House.\24\ The spaces around the White House are regularly used
for First Amendment-protected protests and demonstrations. The possible
use of face recognition to identify individuals near the White House
raises serious First Amendment issues and threatens to chill speech.
---------------------------------------------------------------------------
\24\ USSS PIA at 1, https://www.dhs.gov/sites/default/files/
publications/privacy-pia-usss-frp-november2018.pdf.
---------------------------------------------------------------------------
dhs's use of face recognition lacks proper safeguards and pose
substantial risks
Use of face recognition poses a unique threat to Constitutional
rights.--Participation in society necessarily exposes one's images in
public spaces. But ubiquitous and near-effortless identification
eliminates the individual's ability to control the disclosure of their
identities to others and poses a special risk to the First Amendment
rights of free association and free expression. The proposed plans by
DHS risk creating a world where individuals are forced to submit to
face recognition surveillance simply to exercise their right to travel.
The aggregation of biometric data for the use of face recognition
and the broad dissemination of this data poses cybersecurity risks and
increases the risk of a data breach.--Indeed, a CBP vendor who had
collected images of travelers along with license plate reader data and
other sensitive information was subject to a recent data breach.\25\
---------------------------------------------------------------------------
\25\ Drew Harwell, Hacked documents reveal sensitive details of
expanding border surveillance, Wash. Post (June 21), https://
www.washingtonpost.com/technology/2019/06/21/hacked-documents-reveal-
sensitive-details-expanding-border-surveillance/.
---------------------------------------------------------------------------
Face recognition technology will disproportionately impact already
marginalized groups. Studies have shown that facial recognition has
significantly higher error rates for darker-skinned individuals.\26\ It
is unacceptable for DHS to implement a technology with a documented
racial bias without proving that such a bias has been eliminated.
Moreover, use of face recognition for immigration enforcement raises
further risks of a disproportionate impact on already marginalized
groups.
---------------------------------------------------------------------------
\26\ Joy Buolamwini (MIT Media Lab) and Timnit Gebru (Microsoft
Research), Gender Shades: Intersectional Accuracy Disparities in
Commercial Gender Classification (2018), http://proceedings.mlr.press/
v81/buolamwini18a/buolamwini18a.pdf.
---------------------------------------------------------------------------
The agency continues to expand the use of face recognition beyond
what was ever authorized by Congress.--In fact, the Biometric Entry-
Exit program itself is an example of mission creep. The program
leverages the photos provided by passport applicants to the State
Department, who provided the photos for the specific purpose of
obtaining a passport, only to see those photos used in conjunction with
face recognition technology to create a digital ID. Additionally, the
State Department then disclosed the biometric data to other agencies,
including DHS, and there was nothing a passport holder could do to
prevent the disclosure. And, there is nothing an individual could do to
stop DHS from further disseminating their biometric data.
DHS's use of face recognition lacks the safeguards needed to
prevent overcollection, overly broad uses, wide-spread dissemination,
and unnecessarily long retention.--Moreover, DHS has failed to show
that less invasive alternatives could not be used. DHS has moved
forward with face recognition with a focus on justifying its
implementation and not a focus on whether, given the risks, the
technology should be implemented.
conclusion
Face recognition is an especially dangerous technology in need of
strict limits on its use, robust transparency, oversight, and
accountability. It is imperative that Congress suspend DHS's use of
face recognition until Congress fully debates what, if any, proposed
uses should move forward.
If you have questions, please contact Jeramie D. Scott, EPIC Senior
Counsel, [email protected].
Sincerely,
Access Now
ACLU
American-Arab Anti-Discrimination Committee (ADC)
Algorithmic Justice League
Center for Democracy & Technology
Center for Digital Democracy
Center on Privacy & Technology at Georgetown Law
Constitutional Alliance
Consumer Action
Consumer Federation of America
Council on American-Islamic Relations (CAIR)
Cyber Privacy Project
Defending Rights & Dissent
Demand Progress
Electronic Frontier Foundation
Electronic Privacy Information Center
Fight for the Future
Free Press Action
Freedom Works
Government Accountability Project
Immigrant Rights Clinic
of the University of California at Irvine School of Law
Liberty Coalition
MediaJustice
Mijente
National Immigration Law Center
National Workrights Institute
New America's Open Technology Institute
Open MIC (Open Media and Information Companies Initiative)
Open TheGovernment
Patient Privacy Rights
Privacy Times
Project on Government Oversight
Project South
Public Citizen
Restore The Fourth
TechFreedom
Tri-State Coalition for Responsible Investment
Chairman Thompson. The Chair now recognizes the Ranking
Member of the full committee, the gentleman from Alabama, Mr.
Rogers, for an opening statement.
Mr. Rogers. Thank you, Mr. Chairman. Biometric technologies
have the potential to improve security, facilitate travel, and
better enforce our immigration laws. These technologies range
from facial recognition to fingerprints to DNA. Each of these
methods present unique privacy considerations but also clear
security benefits. Not only does Federal law authorize DHS to
use biometrics to verify identities, it requires CBP to collect
biometric entry and exit data for all foreign nationals. This
requirement has been a long-standing, bipartisan mandate.
Recent technological advancements have finally made it
possible. DHS's primary focus is facial recognition at TSA and
CBP checkpoints, where travelers are already providing IDs to
Government employees. TSOs and CBP agents can review several
hundred IDs in a single shift. As a result, fatigue and human
error allow people with fake IDs to slip into our country every
day. Automating this process with biometric technology will
improve transportation security.
CBP and TSA have done their homework on these checkpoint
pilots and are working to build accurate, effective, and secure
systems. DHS should continue to collaborate with experts at
NIST to ensure they are using accurate algorithms to power
these systems.
Biometric systems advance DHS's mission beyond
transportation security. ICE recently conducted a rapid DNA
pilot program to verify family ties on the Southwest Border. A
90-minute test can replace hours of interviews and document
review. This short pilot found a disturbing number of cases
were men who claimed to be the biological parent of a child
quickly changed their story when asked to submit DNA. The
technology does not store DNA in a central database, and each
machine can be purged daily.
Amid the humanitarian crisis on our border, we should be
looking at things like rapid DNA to protect children from abuse
by smugglers who rent them as a ticket into our country.
Additionally, we should be using biometrics to enforce our
immigration laws. Recent reports have emphasized ICE and FBI's
use of DMV photos to identify criminals. I do not believe that
anyone has a reasonable expectation of privacy in a Government
ID photo. Period. Police have long relied on photo books and
manual photo reviews to identify suspects known as fugitives--
or known fugitives.
Effective facial recognition technologies can improve law
enforcement by ridding this process of bias and human error.
Each of these examples use biometrics as a part of the process.
Technology cannot and should not replace the officer's final
judgment, but it can speed up the identity verification for
millions of people every year.
Halting all Government biometric programs, as some of my
colleagues suggest, is an easy way to avoid hard questions.
Taking the easy way out of this issue will not increase the gap
between technology and our ability to understand it. DHS should
continue to consult with experts at NIST to develop clear
public standards for Government biometric systems. DHS
leadership should ensure that these biometric databases are
secure and have clear privacy guidelines. Congress should
continue to educate itself as we are today, about the way we
can employ this technology responsibly. I thank you, Mr.
Chairman. I yield back.
[The statement of Ranking Member Rogers follows:]
July 10, 2019
Statement of Ranking Member Mike Rogers
Biometric technologies have the potential to improve security,
facilitate travel, and better enforce our immigration laws.
These technologies range from facial recognition, to fingerprints,
to DNA.
Each of these methods presents unique privacy considerations, but
also clear security benefits.
Not only does Federal law authorize DHS to use biometrics to verify
identities, it requires CBP to collect biometric entry and exit data
for all foreign nationals. This requirement has been a long-standing
bipartisan mandate. Recent technological advancements have finally made
it possible.
DHS's primary focus is facial recognition at TSA and CBP
checkpoints, where travelers are already providing IDs to Government
employees.
TSOs and CBP Agents can review several hundred IDs in a single
shift.
As a result, fatigue and human error allow people with fake IDs to
slip into our country every day.
Automating this process with biometric technology will improve
transportation security.
CBP and TSA have done their homework on these checkpoint pilots and
are working to build accurate, effective, and secure systems.
DHS should continue to collaborate with experts at NIST to ensure
they are using accurate algorithms to power these systems.
Biometric systems advance DHS's mission beyond transportation
security.
ICE recently conducted a Rapid DNA pilot program to verify family
ties on the Southwest Border.
A 90-minute test can replace hours of interviews and document
review.
This short pilot found a disturbing number of cases where men, who
claimed to be the biological parent of a child, quickly changed their
story when asked to submit DNA.
The technology does not store DNA in a central database and each
machine can be purged daily.
Amid the humanitarian crisis on our border we should be looking to
things like Rapid DNA to protect children from abuse by smugglers who
rent them as a ticket into our country.
Additionally, we should be using biometrics to enforce our
immigration laws.
Recent reports have emphasized ICE and the FBI's use of State DMV
photos to identify criminals.
I do not believe that anyone has a reasonable expectation of
privacy in a Government ID photo. Period. Police have long relied on
photo books and manual photo review to identify suspects and known
fugitives.
Effective facial recognition technologies can improve law
enforcement by ridding this process of bias and human error.
Each of these examples uses biometrics as one part of a process.
Technology cannot and should not replace an officer's final
judgment. But it can speed up identity verification for millions of
people every year.
Halting all Government biometric programs, as some of my colleagues
suggest, is an easy way to avoid hard questions.
Taking the easy way out of this issue will only increase the gap
between technology and our ability to understand it.
DHS should continue to consult with experts at NIST to develop
clear public standards for Government biometric systems.
DHS leadership should ensure that its biometric databases are
secure and have clear privacy guidelines.
And Congress should continue to educate itself, as we are today,
about the way we can employ this technology responsibly.
Chairman Thompson. Thank you.
Other Members of the committee are reminded that, under the
committee rules, opening statements may be submitted for the
record.
I now welcome our panel of witnesses. Our first witness is
Mr. John Wagner, deputy executive assistant commissioner at the
U.S. Customs and Border Protection.
Next, we have Mr. Austin Gould, assistant administrator for
requirements and capabilities analysis at the Transportation
Security Administration.
Next, we have Mr. Joseph R. Di Pietro, the chief technology
officer of the U.S. Secret Service.
Finally, we have Dr. Charles Romine, the director of the
information technology laboratory at the Commerce Department's
National Institute of Standards and Technology.
I look forward to hearing from you all today.
Without objection, the witnesses' full statements will be
inserted in the record.
I now ask each witness to summarize his statement for 5
minutes, beginning with Mr. Wagner.
STATEMENT OF JOHN P. WAGNER, DEPUTY EXECUTIVE ASSISTANT
COMMISSIONER, OFFICE OF FIELD OPERATIONS, U.S. CUSTOMS AND
BORDER PROTECTION, U.S. DEPARTMENT OF HOMELAND SECURITY
Mr. Wagner. Chairman Thompson, Ranking Member Rogers,
Members of the committee, thank you for the opportunity to
testify before you today on behalf of U.S. Customs and Border
Protection. I would like to begin with a few excerpts from the
9/11 Commission report.
When people travel internationally, they usually move
through defined channels or portals. They may seek to acquire a
passport. They may apply for a visa. They may stop at ticket
counters, gates, and exit controls at airports and seaports.
Upon arrival, they pass through inspection points. They may
transit to another gate to get on an airplane. Each of these
checkpoints or portals is a screening, a chance to establish
that people are who they say they are and are seeking access
for their stated purpose.
The job of protection is shared amongst these many defined
checkpoints. By taking advantage of them all, we need not
depend on any one point in the system to do the whole job. The
challenge is to see the common problem across agencies and
functions, and develop a conceptual framework and architecture
for an effective screening system. Throughout Government and,
indeed, in private enterprise, agencies, firms at the portals
confront recurring judgments that balance security, efficiency,
and civil liberties. These problems should be addressed
systemically, not in an ad hoc, fragmented way. Like I
mentioned, these are excerpts from the 9/11 Commission report.
Before CBP presented our current strategy, airlines,
airports, private vendors, and Government agencies, including
DHS were developing their own independent, biometric-based
schemes. In other words, exactly what the 9/11 Commission
warned against doing--an ad hoc, fragmented approach. CBP has
developed a plan that includes other authorities and
responsibilities in our mission set beyond just the biometric
entry-exit mandate for foreign nationals. We saw the solution
had to encompass the entire travel spectrum. We needed a
solution that would also comport with the modernization and
emerging biometric plans of airports, airlines, and cruise
lines. Why? Well, because we don't have a transportation system
that allows the easy segmentation of only foreign visitors on
international departures.
Previous DHS efforts failed for 10 years because they tried
to create a stand-alone, stovepiped, unintegrated process. As
we all know, those plans were cost-prohibitive, would create
massive congestion, and there was significant opposition from
the airlines and the travel industry. So, as a result, CBP
developed a service that simply automates the manual, facial
recognition process that goes on today when a traveler presents
a passport to establish their identity.
To be clear, CBP is only comparing the picture taken
against photos of previously provided by travelers to the U.S.
Government for the purposes of international travel. This is
not a surveillance program.
Since airlines and cruise lines are already required by
statute to provide the biographic passport details of all
travelers on international itineraries, CBP simply assembles a
small gallery of photos of these expected travelers. These
gallery photos are primarily from passports, visas, and
previous international arrivals. A photo is taken and quickly
searched against these distinct galleries, and thereby
validating the biographic data that has already been vetted for
National security and law enforcement concerns and corresponds
to the traveler we all expect it to. We do not run the photo
taken at the airport or seaport against any other databases or
sources of information if it matches that pre-staged gallery
photo.
If a traveler matches a U.S. passport, then the new photo
taken is deleted. There is no need for us to keep it. U.S.
citizens are clearly not part of the biometric entry-exit
tracking system.
Now, recognizing there have been concerns raised over the
inclusion of U.S. citizens, CBP has existing authorities and
responsibilities to determine the citizenship and identity of
all people traveling internationally. This is a U.S. Government
responsibility, not the private sector. It is also unlawful for
a U.S. citizen to travel internationally without a U.S.
passport.
Now, generally, determination of U.S. citizenship is done
by comparing the traveler against their passport. Again, we are
simply automating and using a computer algorithm to enhance
this manual facial recognition existing process. As we saw at
Dulles Airport a few months ago, we had two travelers
presenting U.S. passports claiming to be U.S. citizens.
However, it was found that they were foreign nationals and
imposters to these documents.
Now, as far as our partnerships with the industry
stakeholders, CBP has developed a standard set of business
requirements that our partners have all agreed to, if their
camera is sending a photo to CBP. The business requirements
clearly stipulate they cannot keep the photos. Going back to
the ad hoc, fragmented approach mentioned earlier, our partners
have voluntarily agreed to the CBP business requirements. This
make a single, simple, consistent, transparent approach to the
use of this technology for international travel.
CBP is already bound by and in compliance with existing
privacy, technology, and data collection requirements found in
the Privacy Act, the E-Government Act and the Homeland Security
Act. Our private-sector partners are basically signing on to
these same requirements. We do recognize we can improve the
public's understanding of these requirements and the opt-out
provisions.
We have published a comprehensive, privacy-impact
assessment, the required system of record notice for our
databases, and rulemaking as commenced to put updates into the
Federal regulations it is currently circulating within the
Government.
In conclusion, we are solving a very difficult challenge:
Biometric exit. We are solving it by focusing on improving the
overall travel experience. We are building a tokenless,
efficient, secure, international travel experience. Airlines
and cruise lines have reported reduced boarding times and
increased passenger satisfaction using this system. This system
will allow us to build a world-class travel system in the
United States. This will be the envy of the world as we try to
keep pace with the record-breaking growth of international
travel. So thank you for the opportunity to be here today, and
I look forward to your questions.
[The prepared statement of Mr. Wagner follows:]
Prepared Statement of John P. Wagner
July 10, 2019
Chairman Thompson, Ranking Member Rogers, and Members of the
committee, thank you for the opportunity to testify before you on the
efforts of U.S. Customs and Border Protection (CBP) to better secure
our Nation by incorporating biometrics into our comprehensive entry-
exit system, and to identify overstays in support of our border
security mission.
CBP has received public support for its use of biometrics from the
International Air Transit Association (IATA),\1\ the World Travel and
Tourism Council,\2\ and the Department of Commerce Travel and Tourism
Advisory Board.\3\ With air travel growing at 4.9 percent per year, and
expected to double by 2031, and an increasingly complex threat posture,
CBP must innovate and transform the current travel processes in order
to handle this new volume without significant personnel and
infrastructure investments. Facial comparison technology will enable
CBP and travel industry stakeholders to position the U.S. travel system
as best in class, which will in turn drive the continued growth in air
travel volume.
---------------------------------------------------------------------------
\1\ https://www.iata.org/pressroom/pr/Documents/resolution-one-id-
agm-2019.pdf.
\2\ https://www.wttc.org/about/media-centre/press-releases/press-
releases/2019/we-must-act-and-assign-priority-and-resources-to-
biometrics/.
\3\ https://www.trade.gov/ttab/docs/
TTAB_Biometrics%20Recommendations%20Letter_042- 919.pdf.
---------------------------------------------------------------------------
As authorized in several statutes and regulations,\4\ CBP is
Congressionally-mandated to implement a biometric entry-exit system.
Prior to the Consolidated and Further Continuing Appropriations Act of
2013 (Public Law 113-6), which transferred the biometric exit mission
from the Department of Homeland Security (DHS) generally to CBP, the
U.S. Government and the private sector were developing independent
biometrics-based schemes. These varied, and often uncoordinated,
investments relied on multiple biometrics and required complicated
enrollment processes.\5\ DHS, the Transportation Security
Administration (TSA), legacy United States Visitor and Immigrant Status
Indicator Technology, and several private-sector companies developed
separate uses for biometrics, creating different guidelines and
business rules, which increased privacy risks and decreased
accountability, as each stakeholder had distinct responsibilities.
---------------------------------------------------------------------------
\4\ The following statutes require DHS to take action to create an
integrated entry-exit system: Section 2(a) of the Immigration and
Naturalization Service Data Management Improvement Act of 2000 (DMIA),
Public Law 106-215, 114 Stat. 337; Section 110 of the Illegal
Immigration Reform and Immigrant Responsibility Act of 1996, Pub. L.
No. 104-208, 110 Stat. 3009-546; Section 205 of the Visa Waiver
Permanent Program Act of 2000, Pub. L. No. 106-396, 114 Stat. 1637,
1641; Section 414 of the Uniting and Strengthening America by Providing
Appropriate Tools Required to Intercept and Obstruct Terrorism Act of
2001 (USA PATRIOT Act), Pub. L. No. 107-56, 115 Stat. 272, 353; Section
302 of the Enhanced Border Security and Visa Entry Reform Act of 2002
(Border Security Act), Pub. L. No. 107-173, 116 Stat. 543, 552; Section
7208 of the Intelligence Reform and Terrorism Prevention Act of 2004
(IRTPA), Pub. L. No. 108-458, 118 Stat. 3638, 3817; Section 711 of the
Implementing Recommendations of the 9/11 Commission Act of 2007, Pub.
L. No. 110-53, 121 Stat. 266, 338; and Section 802 of the Trade
Facilitation and Trade Enforcement Act of 2015, Pub. L. No. 114-125,
130 Stat. 122, 199.
\5\ [sic]
---------------------------------------------------------------------------
In 2017, CBP developed an integrated approach to the biometric
entry-exit system that stakeholders, including other U.S. Government
agencies with security functions, such as TSA, and travel industry
stakeholders, such as airlines, airports, and cruise lines, could
incorporate into their respective mission space. We offered relevant
stakeholders an ``identity as a service'' solution that uses facial
comparison to automate manual identity verification thereby harmonizing
the data collection and privacy standards each stakeholder must follow.
This comprehensive facial comparison service leverages both biographic
and biometric data (which is key to supporting CBP's mission),
fulfilling the Congressional mandate and using the system to support
air travel, and improve efficiency and the efficacy of identity
verification, as stated below.
CBP has been testing various options to leverage biometrics at
entry and departure.\6\ These technologies will make the process for
verifying the identity of individuals for this system more efficient,
accurate, and secure by using facial comparison technology. However,
the use of this technology allows CBP to improve identity verification.
Using data that travelers voluntarily provide, we are simply automating
the manual identity verification process done today. Facial comparison
allows CBP to better identify those who are traveling on falsified or
fraudulent documents, which improves our ability to identify those who
are seeking to evade screening in order to enter the United States,
including those who present public safety or National security threats,
and visitors who have overstayed their authorized period of admission.
Moreover, stakeholders have attested that using biometrics could lead
to faster boarding times, enhanced customer service, better use of our
CBP staffing, and faster flight clearance times on arrival.
---------------------------------------------------------------------------
\6\ Available at: https://www.dhs.gov/publication/dhscbppia-056-
traveler-verification-service-0.
---------------------------------------------------------------------------
CBP has continuously kept Congress abreast of our process through
several Congressional reports, hearings, and briefings. Through the
Consolidated Appropriations Act of 2016 and the Bipartisan Budget Act
of 2018, Congress authorized up to $1 billion in visa fee surcharges
through 2027 to support biometric entry/exit.\7\
---------------------------------------------------------------------------
\7\ Pub. L. 114-113 129 Stat. 2242 (December 17, 2015); Pub. L.
115-123 132 Stat. 64 (February 9, 2018).
---------------------------------------------------------------------------
previous efforts to launch a biometric exit system
Prior to the Consolidated and Further Continuing Appropriations Act
of 2013 (Public Law 113-6), which transferred the biometric exit
mission from DHS to CBP, the U.S. Government and the private sector
were already developing independent biometric solutions.
For example, from January 2004 through May 2007, DHS used kiosks
placed between the security checkpoint and airline gates that would
collect a traveler's fingerprint biometrics. The traveler had the
responsibility to find and use the devices, with varying degrees of
support from the airports where the kiosks were deployed. In 2008, DHS
issued a Notice of Proposed Rulemaking (NPRM) proposing to require that
commercial air and vessel carriers collect biometric information from
certain aliens departing the United States and submit this information
to DHS within a certain time frame. Most of the comments opposed the
adoption of the proposed rule due to issues of cost and feasibility.
Among other things, commenters suggested that biometric collection
should be a purely governmental function, that requiring air carriers
to collect biometrics was not feasible and would unfairly burden air
carriers and airports, and that the highly competitive air industry
could not support a major new process of biometric collection on behalf
of the Government. Additionally, as directed by Congress, from May
through June 2009, DHS operated two biometric exit pilot programs
testing the collection of biometric exit data, first by CBP at the
departure gate using a mobile device, and second by TSA at the security
checkpoint.
DHS concluded from the NPRM comments and pilot programs that it was
generally inefficient and impractical to introduce entirely new
Government processes into an existing and familiar traveler flow,
particularly in the air environment. DHS also concluded that the use of
mobile devices to capture electronic fingerprints would be extremely
resource-intensive. This information helped frame our concept for a
comprehensive biometric entry-exit system that would avoid adding new
processes, utilize existing infrastructure, leverage existing
stakeholder systems, processes and business models, leverage passenger
behaviors and expectations, and utilize existing traveler data and
existing Government IT infrastructure.
cbp's integrated approach to a comprehensive biometric entry-exit
system
Leveraging CBP's current authorities, we are executing
Congressional mandates to test technologies to create an integrated
biometric entry/exit system using facial comparison technology.\8\ This
technology uses existing advance passenger information could be used
along with photographs already provided by travelers to the Government
for the purposes of international travel to create ``galleries'' of
facial image templates to correspond with who is expected to be on an
international flight arriving or departing the United States. These
photographs may be derived from passport applications, visa
applications, or interactions with CBP at a prior border inspection.\9\
Once the gallery is created based on the advance information, the
biometric comparison service compares a template of a live photograph
of the traveler to the gallery of facial image templates. Live
photographs are taken where there is clear expectation that a person
will need to provide documentary evidence of their identity. If there
is a facial image match, the traveler's identity has been verified.
---------------------------------------------------------------------------
\8\ Available at: https://www.dhs.gov/publication/dhscbppia-056-
traveler-verification-service-0.
\9\ U.S. passport and visa photos are available via the Department
of State's Consular Consolidated System. See Privacy Impact Assessment:
Consular Consolidated Database, available at https://2001-
2009.State.gov/documents/organization/93772.pdf.
---------------------------------------------------------------------------
These technologies will make the process for verifying the identity
of individuals for this system more efficient, accurate, and secure by
using facial recognition technology. For technical demonstrations at
the land border, air entry, and some air exit operations, CBP takes
photographs of travelers on CBP-owned cameras. These tests have been
extended on a voluntary basis to exempt aliens \10\ and U.S. Citizens.
Such participation provides facilitative benefits and a more accurate
and efficient method for verifying the identity and citizenship of
these individuals. In other air exit and seaport demonstrations, CBP
does not take the photographs; but specified partners, such as
commercial air carriers, airport authorities, and cruise lines, take
photographs of travelers and share the images with CBP's facial
recognition technology. These partners that deploy their own camera
operators and camera technology must meet CBP's technical and security
requirements. These tests occur on a voluntary basis, and are conducted
consistent with that partner's contractual relationship with the
traveler.
---------------------------------------------------------------------------
\10\ Under 8 CFR 235.1(f)(ii) and 8 CFR 215.8(a)(1), CBP may
require certain aliens to provide biometric identifiers to confirm
their admissibility or, at specified airports, their departure. Some
aliens are exempt from any requirement to provide biometrics,
including: Canadian citizens under section 101(a)(15)(B) of the Act who
are not otherwise required to present a visa or be issued a form I-94
or Form I-95; aliens younger than 14 or older than 79 on the data of
admission; aliens admitted A-1, A-2, C-3 (except for attendants,
servants, or personal employees of accredited officials), G-1, G-2, G-
3, G-4, NATO-1, NATO-2, NATO-3, NATO-4, NATO-5, or NATO-6 visas, and
certain Taiwan officials who hold E-1 visas and members of their
immediate families who hold E-1 visas unless the Secretary of State and
the Secretary of Homeland Security jointly determine that a class of
such aliens should be subject to the requirements of paragraph
(d)(1)(ii); classes of aliens to whom the Secretary of Homeland
Security and the Secretary of State jointly determine it shall not
apply; or an individual alien to whom the Secretary of Homeland
Security, the Secretary of State, or the Director of Central
Intelligence determines this requirement shall not apply.
---------------------------------------------------------------------------
CBP is authorized to require ``in-scope''\11\ aliens to provide
biometric identifiers.\12\ For entry, CBP is using facial comparison
technology with CBP cameras during the inspection process.\13\ For
exit, CBP is operating pilot programs at certain land and sea ports of
entry, and airports using facial comparison technology.\14\ This
technology provides the travel industry with the tools to use facial
comparison to verify traveler identity and transmit information to
CBP.\15\ We have identified best practices from the prior work done by
DHS as well as from our international partners that have informed the
design of a biometric exit system that does not require an inefficient
two-step process or require multiple different biometrics for traveler
identity verification purposes.
---------------------------------------------------------------------------
\11\ ``In scope'' aliens are aliens may be required to provide
biometric identifiers to confirm their inadmissibility, or, at
specified airports, their departure, under 8 CFR 235.1(f)(ii) and 8 CFR
215.8(a)(1).
\12\ See 8 CFR 215.8(f)(ii), 235.8(a)(1).
\13\ Available at: https://www.dhs.gov/publication/dhscbppia-056-
traveler-verification-service-0.
\14\ See 8 C.F.R. 215.8(a)(1).
\15\ Numerous statutes require the advance electronic transmission
of passenger and crew member manifests for commercial aircraft and
commercial vessels. These mandates include, but are not limited to
Section 115 of the Aviation and Transportation Security Act (ATSA),
Public Law 107-71, 115 Stat. 597; 49 U.S.C. 44909 (applicable to
passenger and crew manifests for flights arriving in the United
States); Section 402 of the Enhanced Border Security and Visa Entry
Reform Act of 2002 (EBSVERA), Public Law 107-173, 116 Stat. 543; 8 CFR
217.7; 8 CFR 231.1; 8 CFR 251.5; and 8 U.S.C. 1221.
---------------------------------------------------------------------------
CBP understood the need to build a system that all stakeholders
within the travel continuum could participate in--without building
their own independent system--that could expand to other mission areas
outside of the biometric exit process. To address these challenges and
satisfy the Congressional mandate, we work closely with our partners to
integrate biometrics with existing identity verification requirements
already required, to the extent feasible.\16\
---------------------------------------------------------------------------
\16\ Ibid.
---------------------------------------------------------------------------
The facial comparison technology utilized by CBP is currently able
to match travelers at a rate of greater than 97 percent,\17\ which is
accomplished by comparing against a limited number of faces through the
creation of galleries. Travelers who do not match to the system simply
show their passport documents to a CBP officer or airline gate agent,
and upon confirmation of identity, board the aircraft.
---------------------------------------------------------------------------
\17\ Department of Homeland Security Fiscal Year 2018 Entry/Exit
Overstay Report, available at https://www.dhs.gov/sites/default/files/
publications/19_0417_fy18-entry-and-exit-overstay-report.pdf.
---------------------------------------------------------------------------
While CBP's primary responsibility is National security, we must
also facilitate legitimate trade and travel. The use of facial
comparison technology has enabled CBP to not only address a National
security concern head on by enhancing identity verification but also to
simultaneously improve the traveler experience throughout the travel
continuum. CBP engineered a biometric exit solution that gives CBP,
TSA, and industry stakeholders, such as airlines and airports, the
ability to automate manual identity verification with facial comparison
technology at locations where identity verification is present today.
This may include the departure gates, debarkation areas, airport
security checkpoints, and Federal Inspection Services (FIS) area. CBP
only uses photos collected from cameras deployed specifically for this
purpose and does not use photos obtained from closed-circuit television
or other live or recorded video. As the facial comparison technology
automates the manual identity verification process in place today, it
allows CBP and its stakeholders to make quicker and more informed
decisions.
cbp authorities and regulatory updates
As described above, numerous Federal statutes require DHS to create
an integrated, automated biometric entry and exit system that records
the arrival and departure of aliens, compares the biometric data of
aliens to verify their identity, and authenticates travel documents
presented by such aliens. Most recently, in 2017, Executive Order 13780
called for the expedited completion of the biometric entry-exit data
system.\18\
---------------------------------------------------------------------------
\18\ Numerous other statues require DHS to take action to create an
integrated entry-exit system including: Section 2(a) of the Immigration
and Naturalization Service Data Management Improvement Act of 2000
(DMIA), Public Law 106-215, 114 Stat. 337; Section 205 of the Visa
Waiver Permanent Program Act of 2000, Pub. L. No. 106-396, 114 Stat.
1637, 1641; and Section 414 of the Uniting and Strengthening America by
Providing Appropriate Tools Required to Intercept and Obstruct
Terrorism Act of 2001 (USA PATRIOT Act), Pub. L. No. 107-56, 115 Stat.
272, 353.
---------------------------------------------------------------------------
DHS also has broad authority to control alien travel and to inspect
aliens under various provisions of the Immigration and Nationality Act
of 1952, as amended (INA).\19\ As part of CBP's broad authority to
enforce U.S. immigration laws, CBP is responsible for ensuring the
interdiction of persons illegally entering or exiting the United
States, facilitating and expediting the flow of legitimate travelers,
and detecting, responding to, and interdicting terrorists, drug
smugglers and traffickers, human smugglers and traffickers, and other
persons who may undermine the security of the United States at entry.
CBP also has responsibility to facilitate and expedite the flow of
legitimate travel and trade and detect individuals attempting to
illegally enter or exit the United States.
---------------------------------------------------------------------------
\9\ 8 U.S.C. 1365b mandates the creation of an integrated and
comprehensive system. This statute further provides that the entry and
exit data system shall include a requirement for the collection of
biometric exit data for all categories of individuals who are required
to provide biometric entry data. 8 U.S.C. 1365b(d). As a result, if a
certain category of individuals is required to provide biometrics to
DHS on entry as part of the examination and inspection process, the
same category of individuals must be required to provide biometrics on
exit as well. DHS may require persons to provide biometrics and other
relevant identifying information upon entry to, or departure from, the
United States. Specifically, DHS may control alien entry and departure
and inspect all travelers under 215(a) and 235 of the INA (8 U.S.C.
1185, 1225). Aliens may be required to provide fingerprints,
photographs, or other biometrics upon arrival in, or departure from,
the United States, and select classes of aliens may be required to
provide information at any time. See, e.g., INA 214, 215(a), 235(a),
262(a), 263(a), 264(c), (8 U.S.C. 1184, 1185(a), 1225(a), 1302(a),
1303(a), 1304(c)); 8 U.S.C. 1365b. Pursuant to 215(a) of the INA (8
U.S.C. 1185(a)), and Executive Order No. 13323 of Dec. 30, 2003 (69 FR
241), the Secretary of Homeland Security, with the concurrence of the
Secretary of State, has the authority to require aliens to provide
requested biographic information, biometrics and other relevant
identifying information as they depart the United States.
---------------------------------------------------------------------------
To effectively carry out its responsibilities under the INA upon
both arrival and departure from the United States, CBP must be able to
conclusively determine whether a person is in fact a U.S. citizen or
national, or an alien by verifying that the person is the true bearer
of his or her travel documentation. CBP is authorized to take and
consider evidence concerning the privilege of any person to enter,
reenter, pass through, or reside in the United States, or concerning
any matter, which is material or relevant to the enforcement or
administration of the INA.\20\ A person claiming U.S. citizenship must
establish that fact to the examining officer's satisfaction and must
present a U.S. passport or alternative documentation.\21\ Manual review
of passports has historically been used to carry out this
responsibility, but facial comparison technology can do so with greater
consistency and accuracy.
---------------------------------------------------------------------------
\20\ 8 U.S.C. 1357(b).
\21\ 8 CFR 235.1(b). It is usually unlawful for a U.S. citizen to
depart or attempt to depart from the United States without a valid
passport. See 8 U.S.C. 1185(b); 22 CFR 53.1.
---------------------------------------------------------------------------
CBP is statutorily mandated to fully implement a biometric entry/
exit system, and has clear statutory authority to undertake all
appropriate actions in support of the use of biometrics. To further
advance the legal framework described above, CBP is working to propose
and implement regulatory amendments and will provide progress updates
in the Unified Agenda, as appropriate.
data security
There are 4 primary safeguards to secure passenger data, including
secure encryption during data storage and transfer, irreversible
biometric templates, brief retention periods, and secure storage.
Privacy is implemented by design, ensuring data protection through the
architecture and implementation of the biometric technology.
CBP prohibits its approved partners such as airlines, airport
authorities, or cruise lines from retaining the photos they collect
under this process for their own business purposes. The partners must
immediately purge the images following transmittal to CBP, and the
partner must allow CBP to audit compliance with this requirement. As
discussed in the November 2018 Privacy Impact Assessment,\22\ we have
developed Business Requirements to document this commitment, to which
the private-sector partners must agree as a condition of participation
in the pilots. Unlike with the pilots in the early 2000's, CBP has
established these common system-wide standards (business requirements),
which support CBP's integrated approach to the use of biometrics.
---------------------------------------------------------------------------
\22\ Available at: https://www.dhs.gov/publication/dhscbppia-056-
traveler-verification-service-0.
---------------------------------------------------------------------------
Regarding the recent subcontractor data breach incident, CBP is
very concerned that the unauthorized access of CBP data will undermine
Congressional and public confidence in CBP at a time in which we are
pursuing transformative and innovative initiatives to enhance lawful
trade and travel. We are aggressively investigating the breach of the
subcontractor's systems and potential exposure of traveler and license
plate images. There are two events that are under investigation: (a) A
malicious cyber attack that impacted the systems of a Federal
subcontractor; and (b) the unauthorized access of CBP data by the same
Federal subcontractor.
This incident did not impact any of the air entry/exit partnerships
discussed earlier and is limited solely to certain pilot program data
collected in the land border environment. Airlines are trusted partners
of CBP, given the various statutory airline collection mandates \23\ in
place. Airlines have been reliably providing CBP with advance
electronic transmission of passenger and crew member manifests, as well
as authenticating and verifying the identity of all passengers and
ensuring that the traveling passengers are correctly documented to
enter the receiving country.
---------------------------------------------------------------------------
\23\ Numerous statutes require the advance electronic transmission
of passenger and crew member manifests for commercial aircraft and
commercial vessels. These mandates include, but are not limited to
Section 115 of the Aviation and Transportation Security Act (ATSA),
Public Law 107-71, 115 Stat. 597; 49 U.S.C. 44909 (applicable to
passenger and crew manifests for flights arriving in the United
States); Section 402 of the Enhanced Border Security and Visa Entry
Reform Act of 2002 (EBSVERA), Public Law 107-173, 116 Stat. 543; 8 CFR
217.7; 8 CFR 231.1; 8 CFR 251.5; and 8 U.S.C. 1221.
---------------------------------------------------------------------------
While the data breach investigation is on-going, preliminary
evidence indicates several violations of CBP privacy and security
policies and violation of specific contract clauses. CBP is taking
several actions to ensure the security of CBP systems, to include:
Deploying cyber-enhanced technology (e.g., audit tracking, logging, and
enhanced encryption) to all vehicle lanes to further protect license
plate image data; conducting threat assessments to proactively identify
vulnerabilities; restricting removable media usage and rolling out
enhanced insider threat capabilities; and, updating all contractual,
policy, and security requirements. Additionally, CBP required that the
prime contractor immediately terminate its subcontracting agreement and
its work thereunder. As such, the subcontractor no longer has access to
CBP data.
privacy, transparency, civil rights, and future assessments
CBP is committed to ensuring that our use of technology sustains
and does not erode privacy protections. We take privacy obligations
very seriously and are dedicated to protecting the privacy of all
travelers. CBP complies with all requirements under the Privacy Act of
1974 \24\ (Pub. L. 93-579), as well as all DHS and Government-wide
policies. In accordance with DHS policy, CBP uses the Fair Information
Practice Principles (FIPPs) to assess the privacy risks and ensure
appropriate measures are taken to mitigate any risks from its
collection of data through the use of biometrics. As CBP is bound by
the above-mentioned privacy laws and policies, as well as data
collection requirements, partnering stakeholders are also held to the
same standards, which increases accountability with the use of
biometrics.
---------------------------------------------------------------------------
\24\ 5 U.S.C. 552a.
---------------------------------------------------------------------------
CBP strives to be transparent and provide notice to individuals
regarding its collection, use, dissemination, and maintenance of
personally identifiable information (PII). When airlines or airports
are partnering with CBP on biometric air exit, the public is informed
that the partner is collecting the biometric data in coordination with
CBP. We provide notice to travelers at the designated ports of entry
through both physical and either LED message boards or electronic
signs, as well as verbal announcements in some cases, to inform the
public that CBP will be taking photos for identity verification
purposes and of their ability to opt-out of having their photo taken.
Upon request, CBP Officers provide individuals with a tear sheet
with Frequently Asked Questions (FAQ), opt-out procedures, and
additional information on the particular demonstration, including the
legal authority and purpose for inspection, the routine uses, and the
consequences for failing to provide information. Additionally, in the
FIS, CBP posts signs informing individuals of possible searches, and
the purpose for those searches, upon arrival or departure from the
United States.
Any U.S. citizen or foreign national may notify the airline-
boarding agent that they would like to opt out at the time of boarding.
The airline would conduct manual identity verification using their
travel document, and may notify CBP to collect biometrics, if
applicable.
CBP provides general notification of its biometric exit efforts and
its various pilot programs through Privacy Impact Assessments (PIAs)
and Systems of Records Notices (SORNs),\25\ published at www.dhs.gov/
privacy, and through information, such as Frequently Asked Questions,
readily available at www.cbp.gov. We published a comprehensive PIA
called the ``Traveler Verification Service'' in November 2018, to
explain all aspects of CBP's biometric usage through the program, to
include policies and procedures for the collection, storage, analysis,
use, dissemination, retention, and/or deletion of data.\26\
---------------------------------------------------------------------------
\25\ The SORNs associated with CBP's Traveler Verification Service
are: DHS/CBP-007 Border Crossing Information, DHS/CBP-021 Arrival and
Departure Information System, DHS/CBP-006 Automated Targeting System,
DHS/CBP-011 U.S. Customs and Border Protection TECS. Those SORNs can be
found at https://www.dhs.gov/system-records-notices-sorns.
\26\ Available at: https://www.dhs.gov/publication/dhscbppia-056-
traveler-verification-service-0.
---------------------------------------------------------------------------
The PIA and the public notices specifically highlight that facial
images for arriving and departing foreign nationals (and those dual
national U.S. citizens traveling on foreign documentation) are retained
by CBP for up to 2 weeks, not only to confirm travelers' identities but
also to assure continued accuracy of the algorithms and ensure there
are no signs of bias. As always, facial images of arriving and
departing foreign nationals are forwarded to the IDENT system for
future law enforcement purposes, consistent with CBP's authority. As
U.S. citizens are not in-scope \27\ for biometric exit, photos of U.S.
citizens used for biometric matching purposes are held in secure CBP
systems for no more than 12 hours after identity verification, and are
held for this time period only in case of an extended system outage or
for disaster recovery and are then deleted. We reduced the retention of
U.S. citizen photos to no more than 12 hours as a direct result of
briefings and consultations with Chairman Thompson.
---------------------------------------------------------------------------
\27\ Pursuant to 8 CFR 215 and 235.
---------------------------------------------------------------------------
Additionally, as described above, private-sector partners must
agree to specific CBP business requirements, many of which are outlined
in the recent PIA. CBP is simplifying the information flow to the
traveling public by developing one set of business standards and
privacy guidelines, thereby enabling more comprehension of and
transparency and accountability in the biometric process.
While CBP's commitment to transparency has been demonstrated by the
above efforts, CBP is committed to improving its public messaging and
helping the public better understand the technology. CBP welcomes the
committee's input.
CBP collaborates regularly with the DHS Privacy Office to ensure
compliance with applicable privacy laws and policies, and to build in
privacy protection best practices surrounding CBP's use of biometric
technology. The DHS Privacy Office commissioned the DHS Data Privacy
and Integrity Advisory Committee (DPIAC) to advise the Department on
best practices for the use of facial comparison technology. The DPIAC
published its report on February 26, 2019.\28\ CBP has implemented or
is actively working to implement all of the DPIAC recommendations.
---------------------------------------------------------------------------
\28\ https://www.dhs.gov/sites/default/files/publications/
Report%202019-01_Use%20of%20-
Facial%20Recognition%20Technology_02%2026%202019.pdf.
---------------------------------------------------------------------------
CBP is fully committed to the fair, impartial, and respectful
treatment of all members of the trade and traveling public. CBP has
rigorous processes in place to review data and metrics associated with
biometric entry and exit facial comparison performance to assess and
guard against improper bias. Significant variance in match rates that
can be attributed to demographic variables have not been detected.
Additionally, CBP is partnering with the National Institute of
Standards and Technology (NIST) to conduct a comprehensive analysis of
facial comparison technologies in CBP's biometric entry-exit efforts,
in order to improve data quality and integrity, and ultimately the
accuracy of technology that informs agency decision making that affects
people. NIST will provide guidance and data that allows CBP to set a
threshold, given CBP's security and facilitation goals for large-scale
face recognition of travelers at air, land, and sea POEs.
cbp's progress toward implementing a comprehensive biometric entry-exit
system
Biometric Entry-Exit in the Air Environment
CBP is also enhancing the arrivals process by using facial
comparison technology. With more efficient and more secure clearance
processes, airports, airlines, and travelers benefit from shorter
connection times and standardized arrival procedures. Security is
increased by adding facial comparison as an additional tool to reduce
imposter threat while increasing the integrity of the immigration
system. Since initiating this facial comparison technology in the air
environment on a trial basis, CBP has already identified 6
imposters,\29\ including 2 with genuine U.S. travel documents (passport
or passport card), who were using another person's valid travel
documents as a basis for seeking entry to the United States.
---------------------------------------------------------------------------
\29\ Number of imposters updated as of June 11, 2019.
---------------------------------------------------------------------------
CBP is working toward full implementation of biometric exit in the
air to account for over 97 percent of departing commercial air
travelers from the United States. Stakeholder partnerships are critical
for implementing a biometric entry-exit system, and airports, airlines,
and CBP are collaborating to develop a process that meets our biometric
entry-exit mandate and airlines' business needs. These partnerships
help ensure that biometric entry-exit does not have a detrimental
impact on the air travel industry, and that the technology is useful
and affordable. Stakeholders have attested that using biometrics could
lead to faster boarding times, enhanced customer service, better use of
our CBP staffing, and faster flight clearance times on arrival.
Engagement with additional stakeholders continues on how they can be
incorporated into the comprehensive entry-exit system, and CBP is ready
to partner with any appropriate airline or airport that wishes to use
biometrics to expedite the travel process for its customers.
Biometric Entry-Exit in the Land Environment
In the land environment, there are often geographical impediments
to expanding exit lanes to accommodate adding lanes or CBP-staffed
booths. The biometric exit land strategy focuses on implementing an
interim exit capability while simultaneously investigating what is
needed to implement a comprehensive system over the long term.
Biometrically verifying travelers who depart at the land border will
close a gap in the information necessary to complete a nonimmigrant
traveler's record in CBP's Arrival and Departure Information System,
and will allow us an additional mechanism to better determine when
travelers who depart the United States via land have overstayed their
admission period. Given the limitations outlined above and DHS's desire
to implement the use of biometrics without negatively affecting cross-
border commerce, CBP plans on taking a phased approached to land
implementation.
Facial comparison technology, similar to what is used in the air
environment has been deployed at entry operations at the Nogales and
San Luis, Arizona POEs. CBP plans to expand to additional locations
along the Southern Border in 2019. By using the facial comparison
technology in the land environment, CBP has identified 138 imposters,
including 45 with genuine U.S. travel documents (passport or passport
card), attempting to enter the United States.
Additionally, CBP tested ``at speed'' facial biometric capture
camera technology on vehicle travelers.\30\ From August 2018-February
28, 2019, CBP conducted a technical demonstration of facial comparison
technology on persons inside vehicles moving less than 20 miles per
hour entering and departing Anzalduas, Texas.
---------------------------------------------------------------------------
\30\ Available at: https://www.dhs.gov/publication/dhscbppia-056-
traveler-verification-service-0.
---------------------------------------------------------------------------
Later in 2018, CBP began testing facial comparison technology at
the Peace Bridge in Buffalo, New York in conjunction with the Buffalo
and Fort Erie Public Bridge Authority (PBA) to facilitate the
development of a demonstration project to test the viability of taking
images from moving commercial trucks and comparing them against gallery
images. From fall 2018 to early June 2019, PBA took photographs of
truck drivers and sent them to CBP to assist with calibrating the
project. The development is currently on pause.
Biometric Entry-Exit in the Sea Environment
Similar to efforts in the air environment, CBP is partnering with
the cruise line industry to use facial biometric processing supported
by CBP's biometric comparison service in the debarkation (arrival)
points at seaports.\31\ Facial biometric processing at seaports
replaces the current manual comparison performed by the CBP officer
using the travel document. Automating identity verification allows us
to shift officer focus to core law enforcement functions and reallocate
resources from primary inspections to roving enforcement activities.
Currently, there are 4 sea entry sites and 4 major cruise lines that
are operating facial comparison cameras to confirm the identity of
arriving passengers on closed-loop cruises (which originate and
terminate in the same city). The sea entry sites are Bayonne, New
Jersey; Port Everglades, Florida; Miami, Florida; and Port Canaveral,
Florida. Each cruise line conducting facial debarkation operations
reports that passenger satisfaction feedback to include the debarkation
process is significantly more positive as compared to such feedback
from vessels not using facial debarkation. Engagement continues with
cruise lines and port authorities to expand the technology to other
businesses and locations.
---------------------------------------------------------------------------
\31\ Ibid.
---------------------------------------------------------------------------
conclusion
DHS, in collaboration with the travel industry, is aggressively
moving forward in developing a comprehensive biometric exit system in
the land, air, and sea environments that simply replaces a manual
identity check with facial comparison technology. The traveler is well
aware that their picture is being taken for facial comparison purposes
and more detailed information regarding the program is readily
available to the public. CBP's collaborative biometric efforts directly
addresses the recommendations of the 9/11 Commission Report, which
highlighted that security and protection should be shared among the
various travel checkpoints (ticket counters, gates, and exit controls).
``By taking advantage of them all, we need not depend on any one point
in the system to do the whole job.''\32\
---------------------------------------------------------------------------
\32\ The 9/11 Commission Report at 385-386, available at http://
govinfo.library.unt.edu/911/report/911Report.pdf.
Chairman Thompson. Thank you for your testimony.
I now recognize Mr. Gould for summarize his statement for 5
minutes.
STATEMENT OF AUSTIN GOULD, ASSISTANT ADMINISTRATOR FOR
REQUIREMENTS AND CAPABILITIES ANALYSIS, TRANSPORTATION SECURITY
ADMINISTRATION, U.S. DEPARTMENT OF HOMELAND SECURITY
Mr. Gould. Good morning, Chairman Thompson, Ranking Member
Rogers, and distinguished Members of the committee. Thank you
for inviting me before you today to discuss the future of
biometric identification at the Transportation Security
Administration. I am Austin Gould, the assistant administrator
for requirements and capabilities analysis at TSA. I would like
to thank the committee for working with TSA as we continue to
improve the security of transportation systems and,
particularly, for your support of our officers in the field.
The Aviation and Transportation Security Act of 2001
established TSA and the requirement to screen all passengers
who were boarding aircraft. This screening requirement includes
passenger identity verification. The act specifically mentions
TSA's authority to use biometrics for this purpose. Recognizing
the need to positively identify passengers in an era when
fraudulent means of identification are becoming increasing
prevalent and sophisticated, TSA has consistently sought new
processes and technologies to improve performance while
protecting a passenger's privacy. Biometrics represent such
technology.
In 2018, TSA released a biometrics roadmap, which
identifies the steps that the agency is taking to test and
potentially expand biometric identification capability. The
roadmap has 4 major goals: Partner with Customs and Border
Protection on biometrics for international travelers;
operationalize biometrics for TSA PreCheck passengers;
potentially expand biometrics to additional domestic travelers;
and develop the infrastructure to support these biometric
efforts.
Consistent with the biometrics roadmap, TSA is conducting
pilots that use facial biometrics to verify passenger identity
at certain airports. These pilots are of limited scope and
duration and are being used to evaluate biometric technology
for TSA use. These pilots have been executed in conjunction
with Customs and Border Protection, have been supported by
privacy-impact assessments, and passengers have the opportunity
to not participate. In these cases, the standard, manual
identification process is used.
Last month, I observed the pilot currently under way in
Terminal F in Atlanta for international passengers. The capture
camera used for this pilot was in active mode, meaning that it
only captured a facial image after the passenger was in
position and the officer activated it. After the Committee on
Government Oversight and Reform hearing on 4 June, TSA
collected data in Atlanta that demonstrated that over 99
percent of travelers chose to use biometric identification.
Also, based on feedback from the hearing, we have deployed
signage in both Spanish and English to ensure that passengers
are aware that biometrics are being used and the procedure for
opting out. An example of that signage is currently displayed
on the monitor.
[The information follows:]
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Mr. Gould. TSA is committed to addressing accuracy,
privacy, and cybersecurity concerns associated with biometric
capture and matching. In that regard, and pursuant to section
1919 of the TSA Modernization Act, DHS will submit a report
that includes assessments by TSA and CBP that were developed
with the support of the DHS Science and Technology Directorate.
The report will address accuracy, error rates, and privacy
issues associated with biometric identification.
We will also schedule a meeting with privacy groups later
this summer to ensure that they understand TSA's limited use of
biometric identification, have the opportunity to address any
concerns, and as a follow-on to their participation in TSA's
earlier Biometrics Industry Day.
Looking ahead, TSA plans to continue to build upon the
success of past pilots by conducting additional ones at select
locations for limited durations, to refine requirements for
biometrics use. These pilots will continue to be supported by
privacy-impact assessments, clearly identified through
bilingual airport signage, and passengers will always have the
opportunity to choose not to participate.
Biometrics represents a unique opportunity for TSA. This
capability can increase security effectiveness for the entire
aviation system, while also increasing throughput at the
checkpoint and enhancing the passenger's experience. The
ability to increase throughput while providing more accurate
identification will be essential as passenger volumes continue
to grow at approximately 4 percent annually. In fact, we
experienced our busiest travel day ever last Sunday of the
Fourth of July weekend when we screened approximately 2.8
million passengers and crew.
To close, TSA is systematically assessing biometrics for
TSA use. This identification process will enhance aviation
security while also increasing passenger throughput and making
air travel more enjoyable. TSA's system will only be used for
passenger identification and to direct the passenger to the
appropriate level of screening, automating what is currently a
manual process. It will not be used for any law enforcement
purposes, and as always, passengers will have the opportunity
to not participate.
Thank you for the opportunity to address this important
issue before the committee, and I look forward to answering
your questions.
[The prepared statement of Mr. Gould follows:]
Prepared Statement of Austin Gould
July 10, 2019
Good morning Chairman Thompson, Ranking Member Rogers, and
distinguished Members of the committee. Thank you for inviting me to
testify about TSA's current work on assessing how biometric technology
can potentially improve both the security and efficiency of our
transportation system. In June 2018, I became the assistant
administrator of TSA's Requirements and Capabilities Analysis (RCA)
office. RCA is responsible for driving the strategy and development of
TSA's security architecture and operational capabilities to enhance
security and optimize mission performance through analysis and
innovation. RCA directly supports TSA's mission by assessing current
state operations, conducting gap analyses, managing needs
identification, and developing requirements to generate new and
improved security capabilities in alignment with the future vision of
aviation security.
Assessing biometrics technology for application to TSA's missions
is a key initiative for RCA. I welcome this opportunity to explain to
the committee why TSA evaluates the potential to use facial recognition
technology during its passenger screening process, how TSA leverages
both the work and systems already developed by U.S. Customs and Border
Protection (CBP), and the efforts we have taken to date, and continue
to take, to ensure that cybersecurity, privacy, and civil liberties
concerns are considered and addressed at every stage of biometric
testing and potential deployment.
The U.S. aviation transportation system accommodates approximately
965 million domestic and international passengers annually--this
equates to the screening of roughly 2.2 million passengers, 1.4 million
checked bags, and 5.1 million carry-on bags each day. In fiscal year
2018, TSA screened more than 804 million aviation passengers,
representing a 5 percent volume increase from fiscal year 2017. Despite
the significant progress the U.S. Government has made to improve
transportation security, aviation hubs remain high-value targets for
terrorists. Terrorist modes and methods of attack are more
decentralized and opportunistic than ever before.
To stay ahead of these adversaries, we have to innovate, deploy new
solutions rapidly and effectively, and make the most of our resources.
In enacting the Aviation and Transportation Security Act in 2001,
Congress recognized the importance of having TSA explore the use of
biometric or similar technologies to enhance security in the aviation
domain. As part of its mission to protect the Nation's transportation
systems to ensure freedom of movement for people and commerce, TSA is
exercising this authority to assess the use of biometrics technology,
such as facial recognition, for identity verification, including at the
checkpoint. Our evaluation of the use of biometrics technology is for
the purpose of ascertaining how biometric technology might be used to
automate passenger identity verification processes to fulfill a number
of TSA security requirements, and relatedly, to determine a passenger's
ability to access areas of the airport beyond the checkpoint.
Today, TSA Transportation Security Officers at the Travel Document
Checker position at each checkpoint and airline employees at the check
in desk visually compare the passenger in front of them to their photo
ID to verify identity. TSA seeks to assess whether biometrics
technology can automate these processes in ways that enhance security
effectiveness, improve operational efficiency, and streamline the
passenger experience. TSA's investment in Credential Authentication
Technology (CAT) units provides a key tool through which the agency is
analyzing how biometric facial recognition may be applied and optimized
at the checkpoint. CAT authenticates the security features of a
passenger's identification document and then displays the passenger's
screening status from Secure Flight to ensures that the passenger has
the appropriate flight reservation to progress through security
screening and enter the sterile area. Currently, TSA is assessing the
benefits of adding a front end camera to CAT units to further improve
the identity verification process.
TSA recognized the need to outline a comprehensive approach for how
it might develop and implement biometric solutions. To that end, TSA
issued the TSA Biometrics Roadmap for Aviation Security & the Passenger
Experience, which is available to the public on TSA's website, in
September 2018. The Biometrics Roadmap centers on four goals:
Partnering with CBP on biometrics for international
travelers;
Operationalizing biometrics for TSA PreCheck travelers;
Expanding biometrics to additional domestic travelers; and
Developing support infrastructure for biometric solutions.
Equally important, the Biometrics Roadmap also established as a
guiding principle that TSA will adopt a ``privacy by design'' mindset
that incorporates privacy and civil liberty considerations into each
phase of biometric solution development (design, build, implement). It
also delineates that privacy protections will include restrictions to
prevent the use of biometrics for purposes other than transportation
security unless individuals have opted into other uses. Importantly,
passengers will always have an option to not be processed through
biometrics solutions at our checkpoint.
In 2004 Congress directed CBP to develop a biometric entry/exit
program, and CBP has been developing and deploying an automated facial
recognition solution since 2013 in order to comply with this mandate.
Recognizing the opportunity to align and leverage similar operational
efforts amongst DHS components, TSA signed an agreement with CBP in
April 2018 on the development and implementation of joint work related
to biometric technology at airports. Because of this partnership, TSA
and CBP have collaborated on a series of multi-phased pilots using
CBP's facial recognition technology, the Traveler Verification Service
(TVS), for identity verification at the TSA checkpoint at three major
airports.
The first phase pilot, which TSA conducted at John F.
Kennedy International Airport between October and November
2017, tested TVS's ability to perform facial matching for
volunteer international outbound passengers at the TSA
checkpoint. TSA did not alter any operational procedures during
this phase.
There were 2 second-phase pilot programs, also involving
volunteer passengers. One occurred at Los Angeles International
Airport from August to October 2018, and evaluated using TVS's
facial matching results for passenger identity verification.
The other pilot program, which began in November 2018 at
Hartsfield-Jackson Atlanta International Airport in
coordination with Delta Air Lines, is on-going and testing the
long-term viability of biometrics at check-in, bag drop, and
the checkpoint.
The third phase of pilot programs will focus on TSA's
ability to combine Secure Flight vetting status with the
identification results from TVS's facial matching technology.
This deliberate, iterative approach to assessing facial recognition
technology applications in TSA operations provides the agency with a
significant learning opportunity as well as helping to refine future
testing and pilot designs. We are grounding our exploration of
biometric solutions in rigorous scientific study and analysis as well
as ensuring appropriate privacy and cybersecurity safeguards are in
place. While TSA and CBP coordinate efforts on passenger-facing
biometrics today, TSA is also laying the groundwork for an eventual
transition of relevant storage and matching capabilities to DHS Office
of Biometric Identity Management (OBIM), an entity established by
Congress to provide the Department with enterprise biometric solutions.
TSA has engaged OBIM regularly on the build out of its next generation
Homeland Advanced Recognition Technology, which will modernize and
replace the legacy Automated Biometric Identification System, as well
as to receive the benefit of their subject-matter expertise.
Based on the work of DHS S&T, the National Institute for Standards
and Technology, and other researchers, we are aware of a variety of
concerns related to differences in performance for travelers of
different demographic groups and take this issue seriously. Some of
these concerns pertain to risk of different error rates that correlate
with user race, gender, and age. As required by the TSA Modernization
Act (Public Law 115-254, Oct. 5, 2018), TSA studied matching
performance differences across biometric systems and operational
environments to identify the existence of disparities on these and
other grounds. In fact, pursuant to this Act, TSA will provide a report
to Congress that includes an assessment of these issues.
TSA also recognizes that biometric technologies pose unique privacy
concerns. Reflective of such, TSA continually assesses privacy impacts
and implements, as necessary, various strategies to address them in the
passenger context. Should TSA fully operationalize this technology, it
will mitigate privacy risks through providing robust notice and
meaningful choice of alternatives, ensuring strong data security
measures, deleting biometric data promptly following the passenger
transaction, and focusing the uses of the biometric data to those
directly necessary for transportation security, or as authorized under
the Privacy Act of 1974, 5 U.S.C. 552a. A number of publicly-
available Privacy Impact Assessments (PIAs) on the Traveler
Verification Service (TVS) and CBP's cloud-based facial matching system
have been issued, which TSA has relied upon throughout the
collaboration. These PIAs will be updated and strengthened as necessary
as biometric identification develops further. They can be found on the
DHS Privacy Office's public-facing website for review.
With regard to future endeavors, TSA is committed to protecting
personally identifiable information, being transparent, and proactively
mitigating privacy and civil liberties risks identified in the use of
biometric technology. To that end, the DHS's Fair Information Practice
Principles, known as the FIPPs, which serve as DHS's overarching
privacy principles as applied across the Department, will guide efforts
to protect privacy while achieving the operational and security
benefits of biometrics technology.
Although TSA is still early in its exploration of biometric
technologies, we are excited about the potential security benefits
building this capability may provide. We plan to continue testing and
evaluating biometrics technology in an operational context through
additional pilot programs. TSA is planning for a pilot in the fourth
quarter of this fiscal year at McCarran International Airport to test
the 1:1 matching capabilities of the upgraded front-end CAT machine
with a camera unit for facial recognition procedures in TSA PreCheck
lanes. This pilot will not involve CBP technologies or processes. TSA
is finalizing the PIA for this pilot to ensure the public is aware of
any pilot biometric technology solutions involving the collection,
maintenance, use, or dissemination of personally identifiable
information.
As reflected by TSA's March 2019 Biometrics Industry Day--an event
attended by more than 120 people representing various public and
private stakeholder groups including 5 different privacy advocacy
organizations--we will continue to strive to foster communication,
transparency, and input regarding our findings and approach to
developing biometric solutions. Through the information we obtain from
pilots and stakeholders, we hope to gain a better understanding of the
operational impacts of this technology on travelers and consider that
in developing procedures for the potential use of this technology at
the checkpoint. TSA will continue to work on building a robust
requirements and architecture foundation, develop an acquisition
strategy, and seek to fulfill the goals identified in the Biometrics
Roadmap.
Chairman Thompson, Ranking Member Rogers, and Members of the
committee, thank you for the opportunity to testify before you today. I
look forward to your questions.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Chairman Thompson. Thank you for your testimony.
I now recognize Mr. Di Pietro to summarize his statement
for 5 minutes.
STATEMENT OF JOSEPH R. DI PIETRO, CHIEF TECHNOLOGY OFFICER,
U.S. SECRET SERVICE, U.S. DEPARTMENT OF HOMELAND SECURITY
Mr. Di Pietro. Good morning, Chairman Thompson, Ranking
Member Rogers, and distinguished Members of the committee. I am
Joseph Di Pietro, chief technology officer of the United States
Secret Service. I want to thank you for the opportunity to
appear before you today and to discuss the Secret Service's use
of biometrics in performance of our integrated mission.
As previously conveyed to your committee staff, the Secret
Service has significant concerns about testifying in an open
hearing on how we use facial recognition technology to enhance
our protective mission. Therefore, my testimony today on that
issue will focus on the current facial recognition technology
pilot program we are conducting at the White House complex. The
Secret Service closely guards our means and methods as to how
we execute our protective mission. We are aware that our
adversaries are constantly watching and probing us and could
potentially exploit information discussed in this open
environment to use against us.
It would not be wise or prudent to discuss in a public
setting certain assets, capabilities, and protocols used to
carry out our protective mission. However, we would welcome the
opportunity to provide this information to you in a closed
briefing.
Biometric tools, such as fingerprint analysis and DNA
collection, are used on a regular basis by the Secret Service
to investigate, locate, and sometimes arrest individuals who
have committed crimes, to include offenses related to threats
against Secret Service protectees. We understand that the rapid
expansion of biometric technology creates a need to balance
capabilities with the need to preserve the public's expectation
of privacy, and the Secret Service is committed to ensuring a
balance that protects the rights of all individuals.
With respect to fingerprints and palm prints, the Secret
Service has a long-standing program that plays an integral part
in our investigative and personal security processes. Our
ability to process, store, search, and retrieve fingerprint and
palm print images is an operational necessity.
During the course of investigations involving fingerprint
and palm print evidence, forensics examiners at the Secret
Service utilize a variety of regional and National databases to
search latent prints for matches to known subjects. With
respect to DNA, DNA evidence is one of the most effective
identification tools available to law enforcement today.
Advancements related to DNA technology have been rapid, and the
Secret Service remains dedicated to utilizing new applications
to enhance our integrated mission.
The Secret Service collects DNA samples, along with a
subject's fingerprints, as part of the identification and
arrest process. Samples are sent to the FBI and DNA testing,
search, and storage in the National DNA database.
With respect to facial recognition technology, the Secret
Service recognizes that this technology has the potential to be
a powerful tool that may assist in preventing attacks on our
protectees, and there must be an appropriate balance between
security and any potential privacy or other Constitutional
concerns.
In 2014, former Secretary of Homeland Security Johnson
established an independent protective mission panel to conduct
an assessment of security at the White House complex. Among
other important recommendations, the panel stated technology
systems used on the complex must always remain on the cutting
edge, and the Secret Service must invest in technology,
including becoming a driver of research and development that
may assist in its mission.
In furtherance of these recommendations, the Secret Service
is currently working on a facial recognition pilot. The goal of
the pilot is to determine whether facial recognition technology
could be effectively deployed to enhance our protective
mission. While the pilot started in December 2018 and is
scheduled to be completed by the end of August 2019, the Secret
Service began contemplating this pilot as far back as August
2014.
The participants in the pilot are Secret Service employees
who volunteered to take part in this effort. Designated White
House cameras that are part of the video management system
captured the volunteers as they moved through various locations
around the White House complex. Software running on a server
dedicated to the pilot and on a closed network not connected to
the internet seeks to match the images of the volunteers to the
images in the video streams.
Facial images are stored when associated with a match to
one of the volunteers, and at the conclusion of the pilot, all
images will be purged.
The Secret Service's commitment to maintaining First
Amendment protections and desire to address personal privacy
consideration are central factors behind any future
implementation of facial recognition technology. The Secret
Service will not adopt new technologies unless they have been
thoroughly vetted to ensure that sufficient privacy protections
and data safeguards are in place.
In closing, the protection of our Nation's leaders is
paramount to this agency and to the Nation. The partnerships
represented here today, both in Congress and within DHS, are
critical to the success of Secret Service operations.
I thank you for the opportunity to testify concerning the
agency's use of these evolving technologies, and I look forward
to working with you as we move forward. This concludes my
testimony. I welcome any questions you have at this time.
[The prepared statement of Mr. Di Pietro follows:]
Prepared Statement of Joseph Di Pietro
July 10, 2019
Good morning Chairman Thompson, Ranking Member Rogers, and
distinguished Members of the committee. I am Joseph Di Pietro, chief
technology officer of the United States Secret Service (Secret
Service). I want to thank you for the opportunity to appear before you
today to discuss the Secret Service's use of biometrics in performance
of our integrated mission.
As previously conveyed to your committee staff, we have serious
concerns about testifying in an open hearing on how we use facial
recognition technology to enhance our protective mission. Therefore, my
testimony today on that issue will focus solely on the current pilot
program we have in place at the White House Complex, as outlined in the
Department of Homeland Security (DHS) Privacy Impact Assessment (PIA)
dated November 26, 2018.
Pursuant to Title 18 U.S.C. 3056, the Secret Service is
authorized to protect the President, the Vice President, their
immediate families, and other individuals enumerated in the statute. It
is our responsibility to constantly research and evaluate the benefits
and risks of applying available, new and emerging technologies to keep
our protectees safe and to enhance the capabilities of our front-line
Uniformed Division Officers, special agents, and mission support
employees.
The Secret Service closely guards our ``means and methods'' as to
how we execute our protective mission. It would not be wise or prudent
to discuss in a public setting certain assets, capabilities, and
protocols used to carry out our protective mission. We are aware that
our adversaries are constantly probing us and could potentially exploit
information discussed in this open environment to attack us.
The Secret Service uses biometric tools such as fingerprint
analysis and DNA collection on a regular basis, in accordance with
standards and policies, in order to investigate, locate, and sometimes
arrest individuals who have committed crimes, to include offenses
related to threats against Secret Service protectees.
Facial recognition technology is an effective tool that has the
potential to act as a force multiplier. Accordingly, the Secret Service
seeks to utilize and harness these important advances to enhance our
effectiveness while upholding rights guaranteed by our Constitution.
fingerprint/palm prints
The Secret Service has a long-standing fingerprint and palm print
program that plays an integral part in our investigative and personnel
security processes. The Secret Service's ability to process, store,
search, and retrieve fingerprint and palm print images is an
operational necessity.
The Secret Service Live-Scan Program (SSLSP) is an enterprise-wide
initiative deploying Live-Scan Booking Stations to Secret Service
offices agency-wide. Live-Scan Booking Stations electronically capture,
digitize, and transmit descriptive information, fingerprints, palm
prints, signatures, and photos of both applicants and investigative
subjects who are processed through these stations. The records are
transmitted to the Federal Bureau of Investigation's (FBI) Next
Generation Identification System (NGI) database for an automated search
against over 76 million criminal fingerprint records. Simultaneously,
these records are submitted to the Secret Service's own database for
searching and archiving. The conduit used to forward the information to
the FBI is the U.S. Department of Justice's Joint Automated Booking
System (JABS).
During the course of investigations involving fingerprint and palm
print evidence, forensic examiners at the Secret Service utilize a
variety of regional and National databases to search latent prints for
matches to known subjects. For example, the Secret Service coordinates
directly with the FBI and the DHS via their databases, to include the
DHS Office of Biometric Identity Management's Automated Biometric
Identification System (IDENT).
dna
DNA evidence is one of the most effective identification tools
available to law enforcement today. Advances related to DNA technology
have been rapid, and the Secret Service remains dedicated to utilizing
new applications to enhance our integrated mission. DNA technology can
provide accurate identification, improve prosecution rates, and act as
a deterrent against future criminal acts.
The Secret Service collects DNA samples along with a subject's
fingerprints as part of the identification and arrest process. Buccal
collection kits from the FBI are used during the booking process and
are then returned to the FBI for DNA testing, search, and storage in
the National DNA database.
facial recognition technology
In 2014 former Secretary of Homeland Security Johnson established
an independent Protective Mission Panel (PMP) to conduct an assessment
of the security at the White House complex. Among other important
recommendations, the PMP stated that, ``[t]echnology systems used on
the complex must always remain on the cutting-edge, and the Secret
Service must invest in technology, including becoming a driver of
research and development that may assist in its mission.''\1\
---------------------------------------------------------------------------
\1\ See Executive Summary to Report from the United States Secret
Service Protective Mission Panel to the Secretary of Homeland Security,
2014, p. 7.
---------------------------------------------------------------------------
Facial recognition technology is a significant tool currently being
used with great effectiveness in both the private and Government
sectors. Accordingly, the Secret Service is evaluating the potential
benefits of this technology to this agency's protective mission.
Applied correctly and with appropriate controls, this technology could
potentially be used by the Secret Service to enhance our security
posture at critical protective venues.
Specifically, this technology may have the potential to provide an
early notification to Secret Service personnel of individuals who are
of record with the agency when they approach a protective site. These
individuals would have already made a threat against one of our
protectees or been shown to have expressed an ``unusual interest''
toward one of our protectees and, therefore, pose a serious threat to
protected persons, venues, or the general public in close proximity to
one of our protected sites.
While the benefits of technology associated with facial recognition
may provide greater capabilities than the observations of law
enforcement personnel, the Secret Service is well aware that there must
be an appropriate balance between security and any potential privacy or
other Constitutional concerns. Further, it is noted that the Secret
Service expects to come in contact with thousands of the general public
around the White House every day and that the men and women of the
agency strive to ensure a secure environment while respecting all
individual's Constitutional rights.
facial recognition pilot (frp)
In furtherance of the 2014 PMP report recommendations, the Secret
Service Office of Technical Development and Mission Support is
currently working on a Facial Recognition Pilot (FRP). The goal of the
FRP is to determine whether facial recognition technology could be
effectively deployed to enhance the Secret Service's protective
mission. In addition, the Service will determine whether this
technology would be a fiscally responsible investment that would assist
in identifying subjects of interest to the Secret Service as they
approach a protected site.
While the FRP started in December 2018 and is scheduled to be
completed by the end of August 2019, the Secret Service began
contemplating this pilot in August 2014. Prior to the initiation of the
program, DHS approved and published a Privacy Impact Assessment,
evaluating the privacy risks and associated mitigation strategies.\2\
---------------------------------------------------------------------------
\2\ DHS/USSS/PIA-024 Facial Recognition Pilot (Nov. 26, 2018).
---------------------------------------------------------------------------
The participants in the FRP are Secret Service employees who
volunteered to take part. These individuals had their images loaded
into the FRP server. Video streams capture the volunteers as they move
through various locations around the White House Complex, and images of
the volunteers are matched to the video streams. Subsequently,
volunteers provide notification of their movements in and around the
Complex for comparison with the generated matches in the system.
The video streams feed into both the White House CCTV system and
into the FRP server. The FRP server is operated on a closed network and
is not capable of remote connections. The data collected is stored in a
stand-alone database dedicated only to the pilot testing. Only
individuals cleared by the Secret Service have access to the collection
database, and they are accompanied by agency personnel while accessing
the FRP server. All Secret Service personnel and supporting contractors
with access to the data undergo annual privacy awareness and document
security training. Facial images are stored when associated with a
match to one of the volunteers, and, at the conclusion of the FRP, all
images will be purged.
The data collected throughout the FRP will be evaluated for its
effectiveness and accuracy.
office of biometric identity management (obim)
The Secret Service recognizes the value offered by OBIM and its
biometric data storing, matching, and sharing capabilities to assist
with both our protective and investigative functions. Developing a
partnership with OBIM will provide a valuable means to search, match,
and store our biometric data across DHS components as well as with
external agencies. The Secret Service maintains coordination with OBIM
liaisons and continues to develop capabilities and policies regarding
the use, storage, and dissemination of biometric information.
conclusion
The protection of our Nation's leaders is paramount to this agency
and to the Nation. The partnerships represented here today, both in
Congress and among those of us within DHS, are critical to the success
of Secret Service operations. I thank you for the opportunity to
testify concerning the agency's use of these evolving technologies, and
I look forward to working with you as we move forward.
Chairman Thompson, Ranking Member Rogers, and distinguished Members
of the committee, this concludes my testimony. I welcome any questions
you have at this time.
Chairman Thompson. Thank you for your testimony.
I now recognize Dr. Romine to summarize his statement for 5
minutes.
STATEMENT OF CHARLES H. ROMINE, PH.D., DIRECTOR OF INFORMATION
TECHNOLOGY LABORATORY, NATIONAL INSTITUTE OF STANDARDS AND
TECHNOLOGY, U.S. DEPARTMENT OF COMMERCE
Mr. Romine. Chairman Thompson, Ranking Member Rogers, and
Members of the committee, I am Chuck Romine, director of the
Information Technology Laboratory at the Department of
Commerce's National Institute of Standards and Technology, or
NIST. Thank you for the opportunity to appear before you today
to discuss NIST's role in biometric standards and testing for
facial recognition technology.
In the area of biometrics, NIST has been working with
public and private sectors since the 1960's. NIST's work
improves the accuracy, quality, usability, interoperability,
and consistency of identity management systems and ensures that
United States' interests are represented in the international
arena.
NIST research has provided state-of-the-art technology
benchmarks and guidance to industry and to U.S. Government
agencies that depend on biometrics recognition. NIST leads
National and international consensus standards activities in
biometrics, such as facial recognition technology, but also in
cryptography, electronic credentialing, secure network
protocols, software and systems reliability, and security
conformance testing, all essential to accelerate the
development and deployment of information and communications
systems that are interoperable, reliable, secure, and useable.
NIST's biometric evaluations advance the technology by
identifying and reporting gaps and limitations of current
biometric recognition technologies. NIST's evaluations advance
measurement science by providing a scientific basis for what to
measure and how to measure. NIST evaluations also facilitate
development of consensus-based standards by providing
quantitative data for development of scientifically sound, fit-
for-purpose standards.
Since 2000, NIST's Face Recognition Vendor Testing Program,
or FRVT, has assessed capabilities of facial recognition
algorithms for one-to-many identification and one-to-one
verification.
NIST expanded its facial recognition evaluations in 2017.
NIST broadened the scope of its work in this area to understand
the upper limits of human capabilities to recognize faces and
how these capabilities fit into facial recognition
applications.
Historically and currently, NIST's biometrics research has
assisted the Department of Homeland Security, DHS. NIST's
research was used by DHS in its transition from 2 to 10 prints
for the former US-VISIT Program. Currently, NIST is
collaborating with DHS OBIM on face image quality standards and
with DHS Customs and Border Patrol on the evaluation of their
traveler verification service.
NIST is working with DHS Customs and Border Patrol to
analyze performance impact due to image quality and traveler
demographics, and provide guidance and data that allows CBP to
set a threshold, given CBP's security and facilitation goals
for large-scale face recognition of travelers.
NIST's Face Recognition Vendor Testing Program was
established in 2000 to provide independent evaluations of both
prototype and commercially-available facial recognition
algorithms. Significant progress has been made in algorithm
improvements since the program was created.
NIST is researching how to measure the accuracy of forensic
examiners, matching identity across different photographs. The
study measures face identification accuracy for an
international group of professional, forensic, facial
examiners, working under circumstances approximating real-world
casework. The findings, published in the proceedings of the
National Academy of Sciences showed that examiners and other
human-face specialists, including forensically-trained facial
reviewers and untrained super recognizers, were more accurate
than the control groups on a challenging test of face
identification. It also presented data comparing state-of-the-
art facial recognition algorithms with the best human face
identifiers.
Optimal face identification was achieved only when humans
and machines collaborated. As with all areas for face
recognition, rigorous testing and standards development can
increase productivity and efficiency in Government and
industry, expand innovation and competition, broaden
opportunities for international trade, conserve resources,
provide consumer benefit and choice, improve the environment,
and promote health and safety.
Thank you for the opportunity to testify on NIST's
activities in facial recognition, and I would be happy to
answer any questions that you may have.
[The prepared statement of Mr. Romine follows:]
Prepared Statement of Charles H. Romine
July 10, 2019
introduction
Chairman Thompson, Ranking Member Rogers, and Members of the
committee, I am Chuck Romine, director of the Information Technology
Laboratory (ITL) at the Department of Commerce's National Institute of
Standards and Technology (NIST). NIST cultivates trust in information
technology and metrology through measurements, standards, and testing.
Thank you for the opportunity to appear before you today to discuss
NIST's role in biometrics standards and testing for facial recognition
technology.
biometric and facial recognition technology
Home to 5 Nobel Prizes, with programs focused on National
priorities such as advanced manufacturing, the digital economy,
precision metrology, quantum science, and biosciences, NIST's mission
is to promote U.S. innovation and industrial competitiveness by
advancing measurement science, standards, and technology in ways that
enhance economic security and improve our quality of life.
In the area of biometrics, NIST has been working with the public
and private sectors since the 1960's. Biometric technologies provide a
means to establish or verify the identity of humans based upon one or
more physical or behavioral characteristics. Examples of physical
characteristics include face, fingerprint, and iris images. An example
of a behavioral characteristic is an individual's signature. Used with
other authentication technologies, such as passwords, biometric
technologies can provide higher degrees of security than other
technologies employed alone. For decades, biometric technologies were
used primarily in homeland security and law enforcement applications,
and they are still a key component of these applications. Over the past
several years, the marketplace for biometric solutions has widened
significantly and today includes public and private-sector applications
world-wide, including physical security, banking, and retail
applications. According to one industry estimate, the biometrics
technology market will be worth $59.31 billion by 2025.\1\ There has
been a considerable rise in development and adoption of facial
recognition, detection, and analysis technologies in the past few
years.
---------------------------------------------------------------------------
\1\ https://www.grandviewresearch.com/industry-analysis/biometrics-
industry.
---------------------------------------------------------------------------
Facial recognition technology compares an individual's facial
features to available images for identification or authentication.
Facial detection technology determines whether the image contains a
face. Facial analysis technology aims to identify attributes such as
gender, age, or emotion from detected faces.
nist's role in biometric and facial recognition technology
NIST responds to Government and market requirements for biometric
standards, including facial recognition technologies, by collaborating
with other Federal agencies, law enforcement, industry, and academic
partners to:
research measurement, evaluation, and interoperability to
advance the use of biometric technologies including face,
fingerprint, iris, voice, and multi-modal techniques;
develop common models and metrics for identity management,
critical standards, and interoperability of electronic
identities;
support the timely development of scientifically valid, fit-
for-purpose standards; and
develop the required conformance testing architectures and
testing tools to test implementations of selected standards.
NIST's work improves the accuracy, quality, usability,
interoperability, and consistency of identity management systems and
ensures that United States interests are represented in the
international arena. NIST research has provided state-of-the-art
technology benchmarks and guidance to industry and to Federal agencies
that depend upon biometrics recognition.
Under the provisions of the National Technology Transfer and
Advancement Act of 1995 (Public Law 104-113) and OMB Circular A-119,
NIST is tasked with the role of encouraging and coordinating Federal
agency use of voluntary consensus standards in lieu of Government-
unique standards, and Federal agency participation in the development
of relevant standards, as well as promoting coordination between the
public and private sectors in the development of standards and in
conformity assessment activities. NIST works with other agencies to
coordinate standards issues and priorities with the private sector
through consensus standards developing organizations such as the
International Committee for Information Technology Standards (INCITS),
Joint Technical Committee 1 of the International Organization for
Standardization/International Electrotechnical Commission (ISO/IEC),
the Organization for the Advancement of Structured Information
Standards (OASIS), IEEE, the Internet Engineering Task Force (IETF),
and other standards organizations such as the International Civil
Aviation Organization (ICAO), and the International Telecommunication
Union's Standardization Sector (ITU-T). NIST leads National and
international consensus standards activities in biometrics, such as
facial recognition technology, but also in cryptography, electronic
credentialing, secure network protocols, software and systems
reliability, and security conformance testing--all essential to
accelerate the development and deployment of information and
communication systems that are interoperable, reliable, secure, and
usable.
Since 2010, NIST has organized the biennial International Biometric
Performance Testing Conference; more than 100 biometric experts from
all around the globe traditionally attend. This series of conferences
accelerates adoption and effectiveness of biometric technologies by
providing a forum to discuss and identify fundamental, relevant, and
effective performance metrics, and disseminating best practices for
performance design, calibration, evaluation, and monitoring.
facial recognition tests and evaluations
For more than a decade, NIST biometric evaluations have measured
the core algorithmic capability of biometric recognition technologies
and reported the accuracy, throughput, reliability, and sensitivity of
algorithms with respect to image characteristics such as noise or
compression, and to subject characteristics such as age or gender. NIST
biometric evaluations advance the technology by identifying and
reporting gaps and limitations of current biometric recognition
technologies. NIST evaluations advance measurement science by providing
a scientific basis for ``what to measure'' and ``how to measure.'' NIST
evaluations also facilitate development of consensus-based standards by
providing quantitative data for development of scientifically sound,
fit-for-purpose standards. NIST biometrics evaluations are highly
regarded and valued by developers, users, and policy makers.
NIST conducted the Face Recognition Grand Challenge (2004-2006) and
Multiple Biometric Grand Challenge (2008-2010) programs to challenge
the facial recognition community to break new ground solving research
problems on the biometric frontier. Since 2000, NIST's Face Recognition
Vendor Testing Program (FRVT) has assessed capabilities of facial
recognition algorithms for one-to-many identification and one-to-one
verification.
To better align NIST's evaluation schedule with the pace of facial
recognition advancement in industry and academia, NIST expanded its
facial recognition evaluations in 2017. NIST broadened the scope of its
work in this area to understand the upper limits of human capabilities
to recognize faces and how these capabilities fit into facial
recognition applications. NIST evaluations have quantified accuracy for
investigative-use cases which involve human review of candidates from
an automated system, as well as for fully automated identification
applications in which decisions would be accepted on the basis of an
automated search alone.
NIST's work on demographic effects in facial recognition is on-
going. For example, a report addressing demographic effects in mugshots
collected in domestic law enforcement applications is under development
with an expected publication date of Fall 2019.
NIST provides technical guidance and scientific support for
analysis and recommendations for utilization of facial recognition
technologies to various Federal agencies, including the Federal Bureau
of Investigation (FBI), Office of Biometric Identity Management (OBIM)
at the Department of Homeland Security (DHS), Department of Homeland
Security Science and Technology Directorate (DHS S&T), the Department
of Homeland Security's U.S. Customs and Border Protection agency (DHS
CBP), and the Intelligence Advanced Research Projects Activity (IARPA)
at the Office of the Director of National Intelligence. Further, as DHS
S&T works with Transportation Security Administration (TSA) to
scientifically analyze data from its biometrics pilots to inform TSA's
capability development process, NIST has and will continue to provide
consultation to DHS S&T to assure its analysis methodologies meet
industry standards.
Historically and currently, NIST biometrics research has assisted
DHS. NIST's research was used by DHS in its transition from 2 to 10
prints for the former US-VISIT program and NIST is currently working
with DHS CBP to analyze performance impacts due to image quality and
traveler demographics and provide recommendations regarding match
algorithms, optimal thresholds and match gallery creation for its
Traveler Verification Service program. Currently, NIST is collaborating
with DHS CBP on the evaluation of their Traveler Verification Service
(TVS), and with DHS OBIM on face image quality standards.
nist face recognition vendor testing program
NIST's Face Recognition Vendor Testing Program (FRVT) was
established in 2000 to provide independent evaluations of both
prototype and commercially-available facial recognition algorithms.
These evaluations provide the Federal Government with information to
assist in determining where and how facial recognition technology can
best be deployed. FRVT results also help identify future research
directions for the facial recognition community.
The 2013 FRVT tested facial recognition algorithms submitted by 16
organizations, and showed significant algorithm improvement since
NIST's 2010 FRVT test. NIST defined performance by recognition
accuracy--how many times the software correctly identified the photo--
and the time the algorithms took to match one photo against large photo
data sets.
The 2018 FRVT tested 127 facial recognition algorithms from the
research laboratories of 39 commercial developers and 1 university,
using 26 million mugshot images of 12 million individuals provided by
the FBI. The 2018 FRVT measured the accuracy and speed of one-to-many
facial recognition identification algorithms. The evaluation also
contrasted mugshot accuracy with that from lower quality images. The
findings, reported in NIST Interagency Report 8238,\2\ showed that
massive gains in accuracy have been achieved since the FRVT in 2013,
which far exceed improvements made in the prior period (2010-2013). The
accuracy gains observed in the 2018 FRVT study stem from the
integration, or complete replacement, of older facial recognition
techniques with those based on deep convolutional neural networks.
While the industry gains are broad, there remains a wide range of
capabilities, with some developers providing much more accurate
algorithms than others. Using FBI mugshots, the most accurate
algorithms fail only in about one quarter of 1 percent of searches.
These failures are mostly associated with images of persons with facial
injury and those with a long time lapse (17 years or more for the most
accurate algorithm) since the first photograph. The success of mugshot
searches stems from the new generation of facial recognition
algorithms, and from the adoption of portrait photography standards
first developed at NIST in the late 1990's.
---------------------------------------------------------------------------
\2\ https://www.nist.gov/publications/ongoing-face-recognition-
vendor-test-frvt-part-2-identification.
---------------------------------------------------------------------------
nist face in video evaluation program
The Face in Video Evaluation Program (FIVE) assessed the capability
of facial recognition algorithms to correctly identify or ignore
persons appearing in video sequences. The outcomes of FIVE are
documented in NIST Interagency Report 8173,\3\ which enumerates
accuracy and speed of facial recognition algorithms applied to the
identification of persons appearing in video sequences drawn from 6
different video datasets. NIST completed this program in 2017.
---------------------------------------------------------------------------
\3\ https://www.nist.gov/publications/face-video-evaluation-five-
face-recognition-non-cooperative-subjects.
---------------------------------------------------------------------------
human factors: facial forensic examiners
NIST is researching how to measure the accuracy of forensic
examiners matching identity across different photographs. The study
measures face identification accuracy for an international group of
professional forensic facial examiners working under circumstances
approximating real-world casework. The findings, published in the
proceedings of the National Academy of Sciences,\4\ showed that
examiners and other human face ``specialists,'' including forensically-
trained facial reviewers and untrained super-recognizers, were more
accurate than the control groups on a challenging test of face
identification. It also presented data comparing state-of-the-art
facial recognition algorithms with the best human face identifiers. The
best machine performed in the range of the best-performing humans, who
were professional facial examiners. However, optimal face
identification was achieved only when humans and machines collaborated.
---------------------------------------------------------------------------
\4\ https://www.pnas.org/content/115/24/6171.
---------------------------------------------------------------------------
voluntary consensus standards
When properly conducted, standards development can increase
productivity and efficiency in Government and industry, expand
innovation and competition, broaden opportunities for international
trade, conserve resources, provide consumer benefit and choice, improve
the environment, and promote health and safety.
In the United States, most standards development organizations are
industry-led private-sector organizations. Many voluntary consensus
standards from those standards development organizations are
appropriate or adaptable for the Government's purposes. OMB Circular A-
119 directs the use of such standards by Federal agencies, whenever
practicable and appropriate, to achieve the following goals:
eliminating the cost to the Federal Government of developing
its own standards and decreasing the cost of goods procured and
the burden of complying with agency regulation;
providing incentives and opportunities to establish
standards that serve National needs, encouraging long-term
growth for U.S. enterprises and promoting efficiency, economic
competition, and trade; and
furthering the reliance upon private-sector expertise to
supply the Federal Government with cost-efficient goods and
services.
examples of nist consensus standards development activities
ANSI/NIST-ITL.--The ANSI/NIST-ITL standard for biometric
information is used in 160 countries to ensure biometric data exchange
across jurisdictional lines and between dissimilar systems. One of the
important effects of NIST work on this standard is that it allows
accurate and interoperable exchange of biometrics information by law
enforcement globally and enables them to identify criminals and
terrorists. NIST's own Information Technology Laboratory is an American
National Standards Institute (ANSI)-accredited standards development
organization. Under accreditation by ANSI, the private-sector U.S.
standards federation, NIST continues to develop consensus biometric
data interchange standards. Starting in 1986, NIST has developed and
approved a succession of data format standards for the interchange of
biometric data. The current version of this standard is ANSI/NIST-ITL
1: 2015, Data Format for the Interchange of Fingerprint, Facial & Other
Biometric Information.\5\ This standard continues to evolve to support
Government applications including law enforcement and homeland
security, as well as other identity management applications. Virtually
all law enforcement biometric collections world-wide use the ANSI/NIST-
ITL standard. NIST biometric technology evaluations in fingerprint,
face, and iris have provided the Government with timely analysis of
market capabilities to guide biometric technology procurements and
deployments.
---------------------------------------------------------------------------
\5\ https://www.nist.gov/publications/data-format-interchange-
fingerprint-facial-other-biometric-information-ansinist-itl-1-1.
---------------------------------------------------------------------------
ISO/IEC Joint Technical Committee 1, Subcommittee 37 (JTC1/SC37)--
Biometrics.--From the inception of the ISO Subcommittee on Biometrics
in 2002, NIST has led and provided technical expertise to develop
international biometric standards in this subcommittee. Standards
developed by the Subcommittee on Biometrics have received wide-spread
international and National market acceptance. Documents issued by large
international organizations, such as the International Civil Aviation
Organization for Machine Readable Travel Documents and the
International Labour Office (ILO) of the United Nations for the
verification and identification of seafarers, specify in their
requirements the use of some of the international biometric standards
developed by this subcommittee.
Since 2006, JTC1/SC37 has published a series of standards on
biometric performance testing and reporting, many of which are based on
NIST technical contributions. These documents provide guidance on the
principles and framework, testing methodologies, modality-specific
testing, interoperability performance testing, access control
scenarios, and testing of on-card comparison algorithms for biometric
performance testing and reporting. NIST plays a leading role in the
development of these documents and follows their guidance and metrics
in its evaluations, such as the FRVT.
conclusion
NIST is proud of the positive impact it has had in the last 60
years on the evolution of biometrics capabilities. With NIST's
extensive experience and broad expertise, both in its laboratories and
in successful collaborations with the private sector and other
Government agencies, NIST is actively pursuing the standards and
measurement research necessary to deploy interoperable, secure,
reliable, and usable identity management systems.
Thank you for the opportunity to testify on NIST's activities in
facial recognition and identity management. I would be happy to answer
any questions that you may have.
Chairman Thompson. Thank you very much for your testimony.
I now recognize myself for 5 minutes of questioning.
Mr. Wagner, you talked a little bit about the biometric,
entry-and-exit system, and those of us who have been around, we
have historically supported that system. But in the beginning,
we talked about that system would be only used for foreigners,
and based on what I heard you talk about today, you have
expanded that to take in American citizens. Can you explain the
reasoning for that?
Mr. Wagner. Yes. U.S. citizens are clearly outside the
scope of the biometric entry-exit tracking. The technology we
are using for the entry-exit program we are also using to
validate the identity of the U.S. citizen. Because someone has
to do that. Someone has to determine who is in scope or out of
scope, and someone has to validate that the U.S. citizen is the
person presenting that U.S. passport.
So, once we take the picture and match it against the
passport photo, which is what goes on right now just in a
manual review, we use the algorithm to help make that decision,
and then the photo is discarded after that because there is no
need for us to save it.
Chairman Thompson. Well, and what I am trying to get at is,
this was a policy that CBP more or less expanded even though
Congress gave you the authority to look at foreigners.
Mr. Wagner. Well, it helps us in the airlines determine who
is in scope for biometric exit and who is out, because someone
has to make that determination at the boarding area. It would
be unfair to ask the airline to be able to do that, to
determine who is in scope or out of scope.
Chairman Thompson. But you kind-of see what I am saying,
though. Did CBP come back and say to Congress, we are looking
at expanding this authority, but we need Congressional
approval?
Mr. Wagner. We don't see this as expanding the biometric
entry-exit authority. We see this as using the authorities we
have to determine the citizenship of an individual entering or
departing the United States. If we are looking for a U.S.
citizen departing the United States right now because they have
a warrant for their arrest, we will stop travelers in the jet
way and check their passports.
Chairman Thompson. I understand why you are doing it.
Mr. Wagner [continuing]. Using authorities----
Chairman Thompson. Yes, I understand why you are doing it,
but what I am getting at is part of this hearing is to make
sure that we, as Members of Congress, give you the authority
you need to do your job. But part of what I am hearing is you
have kind-of taken your own initiative to do some things beyond
the scope of authority that Congress gave you. So what I would
like for you to do is provide the committee with the written
policy by which you are doing this.
Mr. Wagner. Yes, absolutely.
Chairman Thompson. Thank you. Dr. Romine--I am going to try
to get it right--you have been advising DHS a lot on some of
these things. Have you looked at this expansion of authority
without Congressional intent with DHS?
Mr. Romine. No, sir. That would be outside of NIST's
mission space, which is technical evaluation and standards of
the algorithms.
Chairman Thompson. All right. Well, have you looked at the
collection of data and how the data management is controlled
once its collected?
Mr. Romine. No, sir.
Chairman Thompson. Mr. Wagner, I am back to you, then.
Explain to the committee, this collection of data that you say
this policy gives you, what do you do with it?
Mr. Wagner. So, when the picture is taken and provided and
comes into CBP and we match it against one of our pre-staged
gallery photos, that is comprised of passports and visas and
previous arrivals, if it is a foreign national, subject to the
biometric entry-exit mandate, that photograph will be sent over
to DHS, to OBIM, to be stored in IDENT, which is the
Department's repository for that information. If it is a U.S.
citizen and that photo matches a U.S. passport or a permanent
resident or somebody outside of the scope of entry-exit, that
photograph would be held for 12 hours and then deleted or
purged from our systems. The only reason we hold it for that
short period of time is just in case the system crashes, and we
have to restore everything.
Chairman Thompson. OK. Are you aware of the recent
subcontractor breach of data?
Mr. Wagner. Oh, yes.
Chairman Thompson. Beg your pardon.
Mr. Wagner. Yes.
Chairman Thompson. So how is that inconsistent with what
you just explained to us?
Mr. Wagner. What we were doing with that subcontractor, is
we were testing their camera on the U.S.-Mexico land border in
a stand-alone pilot system. So it wasn't integrated into the
main CBP network. We were testing the taking of the photographs
and the license plates and the ability to take a picture of a
person in a vehicle and whether that would be matchable. In
this case, apparently the--as far as I understand, the
contractor physically removed those photographs from the camera
itself and put it onto their own network, which was then
breached. The CBP network was not hacked. The contractor--and
what we see is--what I believe is they removed that in
violation of the contract, and that is why a relationship has
been severed with them, and we are conducting an investigation.
Chairman Thompson. So you see my concern about how we
control the data we collect?
Mr. Wagner. Absolutely.
Chairman Thompson. Thank you.
I yield to the Ranking Member.
Mr. Rogers. Thank you, Mr. Chairman.
Mr. Wagner, I want to pick up on what the Chairman is
talking about. My understanding of your response a few moments
ago is that it is your belief that you have the existing
statutory authority to do what you are doing. You are just
exercising new technology in that process. Is that an accurate
representation of what your answer was?
Mr. Wagner. Yes.
Mr. Rogers. Thank you.
Dr. Romine, this has been an evolving technology. Can you
tell us, what have been the big changes, if any, over the last
5 years, when it comes to the use of facial recognition, and
biometrics in general?
Mr. Romine. Certainly. Thank you. The advances have been
dramatic, according to our testing. The accuracy and
capabilities of the newer systems that we have seen in the last
few years----
Mr. Rogers. What would be some examples of newer systems?
Mr. Romine. The advent of convolutional neural networks as
machine learning capability to do the image analysis or image
matching.
Mr. Rogers. Is that AI, is that what you are talking about?
Mr. Romine. It is machine learning and artificial
intelligence, yes, sir.
Mr. Rogers. What else?
Mr. Romine. So these are dramatically improved over
previous technologies that relied specifically on particular
characteristics of faces, for example. With suitable training,
these systems have dramatically improved the accuracy for the
best facial recognition systems.
Now, I want to be clear, for the testing that we have done,
there is still a very wide range of performance in the testing
that we have done, in the algorithms that we have tested, but
the best ones--and we have no direct knowledge of the
convolutional neural networks or the machine learning, because
these are submitted to us as black boxes and we don't examine
that. But in conversations with vendors who have submitted
testing, that is the understanding that we have, is that that
new machine learning capability, that deep neural networks has
been the significant advance.
Mr. Rogers. Has this development or this advancement in the
machine learning alleviated in any way the concerns the
Chairman expressed about facial recognition being less accurate
when it comes to females or darker-skinned individuals?
Mr. Romine. We see, because of the significant increases in
the accuracy across the board, the effect of those demographic
effects is diminishing. We have a report--we are doing an
analysis now, a comprehensive analysis, of demographic effects
under the testing that we have just done, and that report
should be out this fall.
Mr. Rogers. Great. When you have these test results, do you
share those with not only DHS and the agency but the public,
the business community?
Mr. Romine. We do, sir.
Mr. Rogers. Great.
Mr. Romine. We do that through public reporting and also
through dissemination with email and other interested parties.
Mr. Rogers. Do you publish those guidelines for the public
consumption as well?
Mr. Romine. We do. Yes, sir.
Mr. Rogers. Excellent. Mr. Chairman, I have a letter here
from the Security Industry Association supporting the use of
biometrics and facial recognition, and I would like to offer it
for the record.
Chairman Thompson. Without objection.
[The information follows:]
Letter From Don Erikson to Chairman Bennie G. Thompson and Ranking
Member Mike D. Rogers
July 10, 2019.
The Honorable Bennie Thompson,
Chairman, House Committee on Homeland Security, 310 Cannon House Office
Building, Washington, DC 20515.
The Honorable Mike D. Rogers,
Ranking Member, House Committee on Homeland Security, 310 Cannon House
Office Building, Washington, DC 20515.
Dear Chairman Thompson and Ranking Member Rogers: On behalf of the
Security Industry Association (SIA), thank you for holding a hearing on
facial recognition technology. SIA represents over 1,000 companies that
provide safety and security technology solutions vital to public
safety, protecting lives, property, information and critical
infrastructure.
The Security Industry Association (SIA) believes all technology
products, including facial recognition, must only be used for purposes
that are lawful, ethical and non-discriminatory. Advanced image and
video analysis can and should be a catalyst for good. Facial
recognition has proven to be a force multiplier for efforts to protect
the homeland, assist law enforcement, and enhance the mission
capabilities, efficiency, and effectiveness of operations in diverse
ways. However, arbitrary limits will harm Americans who benefit from it
in countless but underpublicized ways.
We are concerned that recent calls to completely ban the use of
facial recognition technology at various levels of government are based
largely on a misleading picture of how the technology works and its
real-world uses in the United States. Such calls misunderstand the role
of accuracy rates in everyday usage of facial recognition systems and
misconstrue the real-world implications when algorithms may not work as
well as intended.
Responsible use of facial recognition technology ensures that
appropriate transparency and accountability measures, stakeholder
education, and privacy considerations and civil liberties protections
are equally taken into account prior to deployment. Further actions may
be needed to reassure the public about how facial recognition
technology is being used and ensure that proper policies are being
followed. However, such actions must be based on sound analysis and
involve input from stakeholders with expertise on the technology.
Prior to considering any legislation impacting the use of facial
recognition technology, we strongly encourage Members to review SIA's
recently published policy paper entitled, Face Facts: Dispelling Common
Myths Associated with Facial Recognition Technology and the U.S.
Department of Justice's Bureau of Justice Assistance Policy Development
Template, which was published in concert with the U.S. Department of
Homeland Security and other law enforcement stakeholders, for law
enforcement use of the technology.*
---------------------------------------------------------------------------
* The attachment has been retained in committee files and is
available at https://www.securityindustry.org/report/face-facts-
dispelling-common-myths-associated-with-facial-recognition-technology/.
---------------------------------------------------------------------------
SIA and our members stand ready to contribute to a constructive
dialog surrounding facial recognition technology. Please let us know if
there is any way we can assist you as you continue to examine these
issues.
Sincerely,
Don Erickson,
CEO Security Industry Association
Mr. Rogers. With that, I yield back. Thank you.
Chairman Thompson. Thank you.
Dr. Romine, just so we are clear, the report you referenced
is not out.
Mr. Romine. That is correct, sir. It should be out this
fall.
Chairman Thompson. So the data right now is that women and
dark-skinned people are misidentified more than anybody else?
Mr. Romine. There are demographic effects that affect age--
so significant changes in age over time--age, race, and sex,
there are demographic effects. Quantifying those in a
statistically valid manner is what we are currently doing.
Chairman Thompson. So is that women and dark-skinned
people?
Mr. Romine. Yes.
Chairman Thompson. OK. Thank you. I am just trying to--
thank you.
The Chair recognizes Mr. Correa for 5 minutes.
Mr. Correa. Thank you, Mr. Chairman, for bringing up this
most important issue. This technology is very interesting
because compared to fingerprints, DNA, you give it without
essentially giving permission. You walk down a corridor, some
camera picks you up, picks up your information, and it is used
without your authority or permission in ways that we don't know
about.
Dr. Romine, you talked about false positives, based on
ethnicity, other factors that are still--that technology has
not gotten to the point where it can account for these factors.
Mr. Wagner, I have a question for you, which is, under the
TSA Modernization Act of last year, it requires a public report
on the deployment of biometric technologies, TSA's assessments
of privacy accurate. That report is now late. Any thoughts of
when that report can be presented to us?
Mr. Wagner. Sure. The report is drafted. It is just
circulating for final approval and signature.
Mr. Correa. So at any time now?
Mr. Wagner. Any time.
Mr. Correa. OK. Will that be something that will be
compared to Dr. Romine's report also that will be coming out
very soon?
Mr. Gould. Sir, moving forward, from a TSA perspective, we
will look at any scientific reports and data that we possibly
can to ensure that biometric identification is performing
optimally for our use cases, yes, sir.
Mr. Correa. So, before we get that report, let me,
nonetheless, ask you, Mr. Wagner, right now, the way facial
recognition is being used by your Department, is that affecting
or unduly burdening foreign travelers, race, gender,
nationality?
Mr. Wagner. No. We are not seeing--in a review of our data,
we are not seeing any significant error rates that are
attributable to a specific demographic. That is why we have
also partnered with NIST to come in and review our data and
help us look at it and make sure.
Mr. Correa. So statistically you do have Mr. Gould, is it,
that is reviewing this data, or who is reviewing this data for
you? To make sure that----
Mr. Wagner. They are----
Mr. Correa [continuing]. What you are saying your
conclusion is that it is not adversely affecting commerce,
tourism? I am from the State of California, where commerce,
tourism, is a big part of our economy. I just want to make sure
we are not having a lot of false negatives.
Mr. Wagner. This is having a beneficial effect on that
because it is allowing airlines and cruise lines to board and
unboard people quicker.
Mr. Correa. Excellent. Want to hear that. Just want to make
sure that we see that in the report.
Mr. Wagner. The passenger experience is being improved by
that. We are reviewing internally our own data, and we are not
seeing noticeable discrepancies in that. But we have partnered
with NIST, and throughout this summer and fall, we will be
examining our data very closely to make sure that we are not
unduly hurting people of a specific demographic.
Mr. Correa. I am glad to hear your enthusiastic, you know,
positive answer that it is not burdening unduly some of those
travelers.
Mr. Wagner. Absolutely. We are not----
Mr. Correa. Because that great Californian Ronald Reagan
said: We got to trust, but we got to verify, too.
Mr. Wagner. Absolutely.
Mr. Correa. So I look forward to seeing your data on that
and making sure we are on it. In terms of the data, the purging
of the data, once you are using it, what system do you have to
audit to make sure that that data is actually purged in a
timely manner? You just mentioned one of your subcontractors
had a breach. That information is somewhere out there. You said
that is the reason you terminated that contract. Yet, to me,
when that information gets out there, terminating a contract is
not enough of a--let's say, a deterrent to making sure that
those kinds of breaches, that data, is actually purged in a
timely manner. Are you doing anything to make sure that we
tighten up that part of your system?
Mr. Wagner. Yes. So the subcontractor may face subsequent
action depending on the results of the----
Mr. Correa. Criminal? Civil?
Mr. Wagner. Potentially.
Mr. Correa. Both?
Mr. Wagner. Potentially. Depending on what the
investigation--and our Office of Professional Responsibility is
investigating this, I believe the IG is investigating this--
depending on the circumstances of how the data was taken and
the intentions and why, you know, how it was used, there
potentially could be criminal actions or----
Mr. Correa. When you have those data breaches, who do you
report those to, and under what time do you actually take to
say, ``Hey, this purge--or this breach happened''?
Mr. Wagner. Well, they are supposed to report it to us
almost immediately. We do report it to Congress if it meets a
certain threshold. Then internally we will----
Mr. Correa. What threshold would that be?
Mr. Wagner. I don't know off-hand.
Mr. Correa. Like to look at that a little closer because
clearly small breach versus a large breach, is that your
threshold, size of the breach? What is the threshold?
Mr. Wagner. I believe it is a hundred thousand, but I will
have to--I will get back to you on that.
Mr. Correa. Mr. Chairman, I am out of time, but I think it
is very important that these kinds of breaches be reported
immediately to Congress.
Chairman Thompson. I agree.
Mr. Correa. Mr. Chairman, I am out of time, so I yield.
Chairman Thompson. Thank you very much. The Chair
recognizes the gentleman from Texas, Mr. McCaul, for 5 minutes.
Mr. McCaul. Thank you, Mr. Chairman. You know, we all want
to protect civil liberties and privacy. When somebody is in the
public domain, as I understood in law school, there is no
expectation of privacy. This technology, in my judgment, has
really protected the Nation from drug smugglers, gang members,
and potential terrorists.
I introduced the BITMAP bill, which is the Biometric
Transnational Migration Alert Program. Last Congress, it passed
unanimously out of this committee. It passed on the House
floor, 272 to 119. Now it is being held up. I would like to
examine what the effect of not authorizing this program would
have.
Mr. Wagner, can you tell me what successes the BITMAP
program has had? Particularly when it comes to individuals
coming from other parts of the world, that are known--that are
basically countries of special interest, special interest
aliens, or KSTs, known or suspected terrorists, coming across,
into this hemisphere up through Latin America, into the United
States of America?
Mr. Wagner. Sure. So the BITMAP Program, it is administered
by ICE. It is a program they work with their foreign
counterparts to utilize fingerprint technology, to take
fingerprints of exactly those populations you just referenced,
as they transit through certain countries in Central or South
America, making their way on up through Mexico to the United
States. So, if they show up in a Central American country, the
foreign authorities will use the BITMAP Program to collect the
passport information and their fingerprints.
When that person ultimately shows up at our Southwest
Border and has mysteriously lost their passport, we are able to
take their fingerprints and match it back up with that previous
encounter in Central America to sufficiently identify who that
person is. This is the passport that they had at that----
Mr. McCaul. Is it true that, through that journey, that
they are--while the names and identities may change----
Mr. Wagner. Sure.
Mr. McCaul [continuing]. Their biometrics do not change?
Mr. Wagner. Correct.
Mr. McCaul. That is the best way to identify who this
person really is?
Mr. Wagner. Correct.
Mr. McCaul. Can you, in this setting--and I don't know if
that is possible--give us some indication of the numbers of
special interest aliens that have been stopped in this program
and also known or suspected terrorists?
Mr. Wagner. Have to get back to you on that. I don't have
any today.
Mr. McCaul. How significant is it?
Mr. Wagner. It is significant. I mean, it is an absolute
vulnerability that, as we have seen, terrorists can exploit,
and it is a vulnerability we need to address.
Mr. McCaul. Dr. Romine, I guess from what I am hearing from
you is that we don't want to get this wrong. I think Mrs.
Watson Coleman was talking about herself being possibly in this
pool of candidates that could get somehow mischaracterized. I
mean, tell us where are we right now with the technology? How
accurate is it?
Mr. Romine. How accurate is it, oh, I see. The very best
algorithms that we have tested the most recently have false
negative rates that are extremely low. The accuracy can range
into the--for the best algorithms in a one-to-many match--can
range into the 99.7 range.
Mr. McCaul. So 99.7 percent accuracy?
Mr. Romine. Accuracy.
Mr. McCaul. That is pretty good.
Mr. Romine. I beg your pardon?
Mr. McCaul. That is a pretty good number.
Mr. Romine. From a scientific standpoint, we report the
number. The judgment on what is a pretty good number is up to
the policy makers, but it is a high number for me.
Mr. McCaul. It is very high. You are a scientist, I am not,
but it sounds pretty high to me. I think it is always a balance
in this committee and when we deal with security issues, you
know, we deal with privacy and civil liberties, we always have
to balance these as Americans, and I think it is important that
we balance those factors. But I wouldn't want to throw the baby
out with the bath water. I think the BITMAP Program has been
extremely successful, has stopped a lot of bad actors from
coming into the United States, and Mr. Chairman and Ranking
Member, I hope that this committee, we could still advance that
authorization and that bill through this Congress because I do
think it is important to protect the American people. It is one
of the most important responsibilities that we have as Members
of Congress.
With that, I yield back.
Chairman Thompson. Thank you very much.
The Chair recognizes the gentle lady from New Mexico, Ms.
Torres Small, for 5 minutes.
Ms. Torres Small. Thank you, Chairman. Last month, the CBP
announced that there was a data breach with some of the
subcontractors operating at land ports of entry along the
Southern Border, and as a result, thousands of license plate
numbers and images of drivers were taken, and images of drivers
that were taken by facial recognition technology were
compromised. I represent multiple border towns, where you cross
back and forth into Mexico for jobs, shopping, tourism,
medicine.
Also, within the interior of the district, there are border
checkpoints, and when they are operating, that same information
is being taken--license plates and facial--and pictures of
people's faces. So we want to be able to make sure that the
citizens' data is secure. Were there audits into the
subcontractor's system prior to the hack?
Mr. Wagner. I am not aware of that. I don't know. I will
have to check.
Ms. Torres Small. Can you get back to us on that, please?
Thank you.
And did these private subcontractors have the authority to
store those U.S. citizens' data?
Mr. Wagner. They did not have the authority to have the
pictures taken by the camera, from what I understand.
Ms. Torres Small. Oh, so not even to store it, they did not
have the authority to take any pictures of faces?
Mr. Wagner. They had the authority to take them. They did
not have the authority to take it off the camera and put it
onto their own network, which is apparently what happened.
Ms. Torres Small. They did. OK. What protocols did CBP have
in place to oversee contractor and subcontractor data security
practices?
Mr. Wagner. I mean, they go through background checks. They
are vetted. They are cleared. They are trained on use of the
systems that they are going to work on. As far as having the
audit controls on--this was a stand-alone pilot, so it was
outside of our normal network, and we apparently did not have
the same level of controls and audit capabilities on that,
because it was a stand-alone, closed system. Those are things
being put into place now on all those systems to make sure you
can't connect a portable media drive on that and extract
information. You know, our main network has these protocols on
them, but we didn't have them on this type of system.
Ms. Torres Small. So did you say those are in place now?
You have corrected the problem?
Mr. Wagner. They are being put into place now.
Ms. Torres Small. They are being put into place now?
Mr. Wagner. Yes.
Ms. Torres Small. Can you follow up and let us know when
they are in place?
Mr. Wagner. Absolutely, yes.
Ms. Torres Small. Because that is something of deep
concern. Thank you.
With all pilot programs, because I remember going through
the border checkpoints and being told, you know, this is a
pilot, so don't worry about it yet----
Mr. Wagner. Right.
Ms. Torres Small. It is just a pilot. That is actually when
we need to make sure that we are operating it correctly.
Mr. Wagner. Agreed.
Ms. Torres Small. So I want to switch now to Congressional
authorization.
Mr. Wagner, it is my understanding that it is the law that
Congress is enacting a biometric entry-exit system limit data
collection to foreign nationals. Is that correct?
Mr. Wagner. Yes.
Ms. Torres Small. OK. Under what authority is CBP
collecting biometric information on U.S. citizens as part of
the entry-exit system?
Mr. Wagner. We are using the information under 8 U.S.C.
1357B and 8 CFR 235.1, which allows us to consider any
information or evidence pertaining to a person crossing the
border in establishing their U.S. citizenship. So, generally, a
person will present a U.S. passport to us. We can look at it.
We can manually review. We can ask them questions how they
obtained it.
Ms. Torres Small. Thank you. Actually I am going to switch
direction really quickly. I apologize. I know some of that was
already covered, so I appreciate it.
I want to switch to the Federal agencies that are scanning
through U.S. citizens' driver's licenses, and ICE is one of
those that has been identified as potentially scanning through
these databases. For what purpose--or are your components
currently attempting to or successfully accessing State
driver's license databases in any way?
Mr. Wagner. So, for the biometric program we are
discussing, we are not using driver's license information. We
do use driver's license information from the States that have
entered into agreements with us, where their driver's license
also substitutes for a passport to cross the border. I think we
have about 5 U.S. States and maybe 4 Canadian provinces that
entered into written agreements with us to mark the citizenship
of the driver's license holder on the document, so they can
cross the border without having to go get a passport. That
serves in lieu of the passport.
Ms. Torres Small. Does the DMV in those States require
probable cause or warrants to access that information?
Mr. Wagner. Well, when that person crosses the border, our
agreement allows us to verify with them that that is a valid
license and to retrieve the photo from that so we can see who
it belongs to. We also have other law enforcement access
through--into biographical driver's license data that we also
might use in a law enforcement context that is very common for
law enforcement agencies to access.
Ms. Torres Small. Thank you, Mr. Wagner.
Chairman Thompson. Thank you very much.
The Chair now recognizes the gentleman from New York, Mr.
Katko, for 5 minutes.
Mr. Katko. Thank you, Mr. Chairman, and thank you all for
being here today.
Just take a step back for a moment. As a Federal prosecutor
for 20 years routinely dealing with homicides and matters of
violent crime, some of the tools in the toolbox I had were
fingerprints at first and, later, DNA.
When they both came on-line, at first there were concerns
about how they would be used, and now they are becoming more
mainstream. I hope and pray that it is the same with facial
recognition.
But, you know, all three have the capability not only of
helping to solve crimes but also making sure that crimes aren't
committed. But even something we don't think about enough is
exonerating people who are falsely accused. I mean, look at
what the DNA system has done for people falsely accused in
prisons. It has been a remarkable breath of fresh air.
So my concern is not with the efficacy of using it. My
concern is that we get it right. Like we have done with
fingerprints and like I think we are doing with DNA.
So my questions focus on the accuracy and the things we
need to do to make it better. My colleagues have asked some
great questions about the use of it and the extent of the use,
and we are going to have to have more discussions about that. I
am very concerned about the accuracy.
That was a big thing with DNA starting out, and now DNA
is--the accuracy in the testing is amazing. It is almost--it is
dispositive almost all the time. I don't think we are there yet
with facial recognition. I would like to get there.
So, with that in mind, I want to ask Mr. Romine a couple of
questions. You talk about the fact that you are charged with
examining the gaps and limitations of certain things, including
facial recognition.
So what do you see as the gaps and limitations of it right
now?
Mr. Romine. The principal gaps and limitations we see
involve a couple of things. One is image quality. It is still
true garbage-in/garbage-out for software systems. So image
quality has a huge impact.
We see--as I said, I will have a report on demographics,
and there are certain issues associated with demographic
effects. That is particularly true when you are trying to
identify someone when you have a reference image that is maybe
10, 20 years earlier than the person that you are trying to
identify. That can be a very big challenge.
Similarly, if someone has been injured or there is some
obscuring of the face for other reasons, that can have a
challenge.
Images that are taken noncooperatively. I don't mean
uncooperative. I mean, where someone is not standing still
looking at a camera with the intent of registering an image. If
you are taking an image through a windshield, for example, or
if you are taking an image of someone who is walking and not
facing a camera, those can have a significant impact on the
accuracy and the ability of these systems to do identification.
Mr. Katko. OK. What can we do to improve that portion of
it?
Mr. Romine. The industry continues to make advances. I
mentioned the emergence of convolutional neural networks as a
game-changer in this space. I think we don't know what we don't
know coming down the pike, but I think there continue to be
improvements that we see in our testing over time. So the
industry is making great strides.
Mr. Katko. You mentioned also, in response to a question
from one of my colleagues, that the demographic effects of
facial recognition software are diminishing.
Could you expound on that and what you mean by that?
Because you say it is 99.7 percent accurate. But it is probably
not 99.7 percent accurate for certain segments. So like, for
example, darker-skinned female, I want to know what you are
doing to make that better and how we can make it stronger.
Mr. Romine. That is correct. From NIST's perspective, what
we do to make things better is provide an evaluation
capability. So we are not doing any training----
Mr. Katko. That is understood.
Mr. Romine [continuing]. Development. However, I would say
that anytime the overall performance of the system improves as
dramatically as facial recognition has improved over the last 5
to 6 years, the compression--the effect of differences in
demographics shrinks as well. And the report later, once we
have finished our analysis, the report that comes out in the
fall, will----
Mr. Katko. That sort-of answers my question. But you admit
that certain demographics have a disproportionate error rate.
So you are saying it is improving. How much has it improved?
Mr. Romine. We haven't finished the analysis yet, so I am
not able to answer that question currently. The report will
come out in the fall.
I will say that the--it is unlikely that we will ever
achieve a point where every single demographic is identical in
performance across the board, whether that is age, race, or
sex. But we want to know just exactly how much the difference
is.
Mr. Katko. This report will detail that when it comes out
in the fall?
Mr. Romine. Yes, sir.
Mr. Katko. All right. Thank you very much. I yield back,
Mr. Chairman.
Chairman Thompson. We all look forward to the report.
Mr. Katko. Indeed.
Chairman Thompson. I assure you.
The Chair now recognizes the gentlelady from Illinois, Ms.
Underwood.
Ms. Underwood. Thank you, Mr. Chairman.
Mr. Gould, I represent Illinois' 14th District, where we
drive about an hour or two to get to major airport in Chicago.
So our community is always interested in learning more about
the technologies that can potentially improve security at
airports while still reducing the flier's wait time.
However, before implementing any new technologies, like
biometric screening, it is really important, crucial even, to
make sure that they are proven to be effective, reliable, and
fair.
So can you please run through the ways in which TSA is
currently employing biometric screening at checkpoints?
Mr. Gould. Yes, ma'am.
Currently, we are only using biometrics technology in the
international terminal, Terminal F, in Atlanta. That is on a
pilot basis.
Our approach to biometrics implementation at TSA is
extremely deliberative. We want to understand how the
technology works, how it can improve identity verification for
the traveling public, and how it can improve the passenger
experience.
Going back to the discussion on image quality that happened
before, we are in a fortunate case at TSA in that we really
control the environment in our checkpoints so we can ensure
optimal lighting, optimal distance from the camera, so we get
the highest quality images possible for biometric matching. For
the pilot in Atlanta, we are matched up with CBP using their
TVS system, and we see extremely high match rates there.
Moving forward, we will look to pilot 1-to-1 matching
capability where a traveler will provide a credential, that
credential will be assessed by our CAT machine, and it will
return a match rate on whether or not the face that has been
captured matches the face that is embedded in that credential.
In that scenario, no information even leaves the checkpoint
and nothing is retained on the camera. So that is some of the
things that we are looking at. I believe that, when we are
through with, you know, these pilots that we are doing for
biometric development, we will see that we cannot only improve
passenger security but also make it a much more positive
experience for the traveling public by reducing wait times.
Ms. Underwood. That is great.
How are the airports and airlines using the biometric
security screening technology beyond the TSA checkpoints, if
you are aware, and what other uses are planned for the future?
Mr. Gould. So, right now, I can comment on really what we
are doing in Atlanta with Delta Airlines.
Ms. Underwood. OK.
Mr. Gould. In Atlanta, the Delta Airlines kiosks use
biometric identification when the passenger checks in, to make
sure--should they choose to do so--to make sure that that
person is the passenger who is ticketed on that particular
flight.
TSA has oversight of the bag drop to ensure the passengers
are positively matched to bags in the international--you know,
for international travel. So Delta Airlines has a security
program amendment that we have granted them to use biometric
technology to do that matching at the bag drop.
We use it at our checkpoint in Atlanta. Then it is, of
course, subject--it is used at the exit point, at the gate.
Ms. Underwood. OK. So is that the only specific agreement
with an airport or airline that TSA has to govern the use of
biometrics? So you said----
Mr. Gould. Right now, the security program amendment that
we have granted Delta for the limited use only in Atlanta is
the only formal agreement that we have entered into with the
airlines.
Ms. Underwood. So does TSA have any role in improving
airport and airline uses of biometric technology?
Mr. Gould. We have roles in improving the use of biometric
technology where TSA has equities. Again, I would go back to
say that would be the checkpoint and the bag drop. So, if an
airline wanted to use biometrics at the bag drop to positively
match that traveler to that bag, they would have to request a
security program amendment, and we would have to issue it.
Ms. Underwood. OK. As the use of biometric data continues
to expand, Illinoisans understandably have a lot of questions
about how such sensitive personal data is used and stored.
So I would like to open this question up to the panel.
Under what circumstances do your components collect
biometric data on U.S. citizens?
We can start with Mr. Wagner.
Mr. Wagner. You say collect on U.S. citizens?
Ms. Underwood. Yes, sir.
Mr. Wagner. We are temporarily holding it while we validate
that it corresponds to the passport that person is presenting,
and then it is purged after 12 hours from our system.
Ms. Underwood. OK.
Mr. Gould. From a TSA perspective, we are leveraging
photographs that travelers have provided to facilitate travel
like passport photographs. When we capture the image at the
checkpoint, it is not retained at the camera. Once that image
is encrypted and transmitted, we only get back a match result.
Ms. Underwood. Interesting. OK.
Mr. Di Pietro. Ma'am, Secret Service collects fingerprints,
palm prints, mugshots, other identifying information on
individuals who we arrest as part of our criminal
investigations.
Ms. Underwood. But not as part of regular screening?
Mr. Di Pietro. Pardon?
Ms. Underwood. You don't retain the data that you collect
as part of the regular screening?
Mr. Di Pietro. That is correct.
Ms. Underwood. You don't store it?
Mr. Di Pietro. No, no. Regular screening, we use metal
detectors, cabinet X-rays, things like that cap.
Ms. Underwood. And fingerprints. So to get into the White
House----
Mr. Di Pietro. No, we do not use fingerprints at the White
House. We don't scan for that.
Ms. Underwood. Great.
Yes, sir?
Mr. Romine. The data that we have is sequestered in servers
that are air gapped--they are not connected to the internet--in
a locked door. I am the director of the laboratory, and I am
not permitted to go into that room without being escorted. So
it is very tightly controlled.
Ms. Underwood. Thank you so much.
I yield back.
Chairman Thompson. Thank you.
The Chair recognizes the gentleman from North Carolina, Mr.
Walker.
Mr. Walker. Thank you, Mr. Chairman.
Ninety-nine-point-seven percent, that is pretty good, or
about ironically the same on-base percentage that Cedric
Richmond has at our annual baseball game, but that is another
topic. A problem, I should, say not a topic.
But I do have a question for you, Dr. Romine.
How do you ensure--and I think Ms. Underwood was just kind-
of approaching this. How do you ensure that the biometric data
collected is secured?
Let me unpack a little bit more. Is the biometric
identifier connected directly to other possibly sensitive or
private information about the person?
Mr. Romine. The data that we have on facial recognition is
not connected to identifying information. So I will have to
double check the exact features there, but I am----
Mr. Walker. Can you do that for us and report back?
So you are saying that the information that you are
collecting is secured?
Mr. Romine. The information we are collecting--we don't
collect information. We obtain it from our partners for the
purposes of evaluation only, and we secure that in--it is in a
secure server.
Mr. Walker. Let's use the word ``obtain'' instead of
``collect.''
Have you ever had a breach on the information you have
obtained?
Mr. Romine. No, sir.
Mr. Walker. OK. Thank you.
Questions for the panel. Keep it about 10 or 15 seconds
will be good. That way we can get everybody in here.
Can you elaborate more on these programs that have been
successful, specifically on the ones identifying facial
recognition, any other biometric technologies? If you can
elaborate either on the success of them or adding security
benefits or expediting travel for passengers?
We will start with Mr. Wagner.
Mr. Wagner. Sure. It gives us the ability to validate a
person's biographical identity within 2 to 3 seconds without
having to handle the physical passport and allows us to link it
up in a secure way. So the person we did all our National
security checks against in TSA, international security checks
on international flights, corresponds to the person who is in
front of us.
Mr. Walker. Mr. Gould.
Mr. Gould. Sir, with our pilot in Atlanta, we do data
collection on the number of people who were choosing not to
provide biometric identification at our checkpoint, and it was
less than 1 percent. People seem to enjoy it. The traveling
public moves through the checkpoint very rapidly. The best part
of it is we enhance identity verification, thereby enhancing
security.
Mr. Walker. OK. Mr. Di Pietro, does it impact you at all?
Mr. Di Pietro. Not really. Right now, we are piloting some
technology, but we are in the middle of that test right now, so
we haven't compiled the data. The tests will finish up at the
end of August, and then we will have a chance to go through and
review the data, and then we will be able to draw some
conclusions. But at this point, we are still in the middle of
the test.
Mr. Walker. Dr. Romine, anything there?
Mr. Romine. No, sir.
Mr. Walker. OK. Going down the panel again.
Based on these successes--specifically Mr. Wagner and Mr.
Gould--where do you see the use of biometric technologies
expanding in your specific agency, even beyond a complete roll-
out of the pilot programs?
Mr. Wagner.
Mr. Wagner. It will significantly transform the arrivals
and departures on international travel in all our different
environments, air, land, and sea, and can really build a very
convenient, efficient, facilitative but yet secure process for
us to do that.
Mr. Walker. Mr. Gould.
Mr. Gould. Sir, so for us, we will build on the success of
our international partnership with CBP that we are doing in
Atlanta to other international travel locations. We will look
to use the CBP system for our trusted traveler population--
PreCheck, Global Entry--to do one-to-few or one-to-many
matching for biometrics purposes at our checkpoints.
Then really the next step that we are looking at is that 1-
to-1 matching that I mentioned before, where a traveler can
approach the checkpoint, provide a credential, have the CAT
machine, credential authentication technology machine, assess
the image embedded in that credential and then match it to a
photograph that is taken right there.
Mr. Walker. All right.
Mr. Di Pietro, do you ever share your information with
local or State governments?
Mr. Di Pietro. Information with respect to fingerprints?
Mr. Walker. Information that you collect. Let me back up
and ask this question, because I think I have got time to get
it in. Ms. Underwood asked a couple of questions, and there
seemed to be just a touch of hesitancy, so I wanted to follow
back up there.
The data that you collect, is it ever collected without
subjects being aware?
Mr. Di Pietro. No, sir.
Mr. Walker. OK. All right. So the information that you do
collect, fingerprints, et cetera, do you ever share that with
State or local?
Mr. Di Pietro. I would have to check with our lab director
on that, sir, and get back to you.
Mr. Walker. Are you familiar with any circumstances that
you have in the past?
Mr. Di Pietro. Sir, I am the Secret Service's chief
technology officer. I work more on the engineering and
technical side. I would have to get with our Forensic Services
Division to answer that.
Mr. Walker. Fair enough. I thank you for your time.
I yield back, Mr. Chairman.
Chairman Thompson. Just let me comment.
In a Classified setting, we are going to ask that question
again of the data collected that people don't know, because I
think there is information being collected in the pilot at the
White House that is different from the answer. But we plan to
have a Classified briefing on that issue.
The Chair recognizes the gentlelady from New York for 5
minutes, Ms. Clarke.
Ms. Clarke. Thank you, Mr. Chairman.
Some would say let's not make--when it comes to National
security, let's not make the perfect be the enemy of the good.
But, unfortunately, the good is not good enough when bias is
baked into the algorithms that create false positives. The
stakes are far too high for individuals and too costly,
particularly for women and people of color.
The wide-scale deployment of facial recognition technology
will have profound implications on privacy. We must look before
we leap. It is imperative that Congress impose safeguards
against mission creep and ensure biased algorithms do not make
their way into wide-spread use.
As a New Yorker, one who lives just miles away from Ground
Zero, National security is crucially important. I know that
first-hand.
But facial recognition technology that routinely
misidentify women and people of color don't make us safer; they
make us less safe. Using this technology to help ICE target
immigrants for deportation doesn't protect us from terrorism;
it terrorizes hardworking families. When CBP uses these
technologies on U.S. citizens traveling abroad without
providing a transparent opt-out process, that is potentially
unlawful.
We have seen what happens when technology is widely
deployed before Congress can impose meaningful safeguards.
Let's not make the same mistake with facial recognition
technology.
You have a contractor that has a breach, and we know that
we are seeing more use of video; deep-face, if you will. That
information gets in the hand of an adversary overseas, and they
want to create a disruption in our Nation, all you have to do
is take that information, create a video from it, and, bam, we
are already into a really bad situation.
I don't know if we are looking at the interconnectedness of
all of these technologies, particularly because they are all
evolving. I am very concerned about the lack of specificity
that we have at this stage.
So my question is about accuracy. Mr. Wagner, CBP boasts
that the facial recognition algorithm it uses is able to make a
match of 98 or 99 percent of the time. But that statistic does
not include instances where facial recognition technology is
unable to capture a high-quality image due to human error, poor
lighting, or other environmental factors.
Recent testing by the DHS science and technology director
has shown that, when data capture factors are included, the
error rate increases to around 10 percent.
Do you dispute S&T's findings?
Mr. Wagner. No.
Ms. Clarke. OK. Why does CBP insist on tracking a bogus
statistic that ignores passengers who cannot be photographed
well enough by the system to be matched?
Mr. Wagner. What we are accounting for is, if we take a
photograph that is of sufficient quality, are we able to match
it.
Ms. Clarke. If?
Mr. Wagner. Correct.
Ms. Clarke. OK.
Mr. Wagner. Then we know we need to address the camera
itself and the lighting conditions to make sure that we are
capturing 100 percent of those photographs that we can then
match at the 98 to 99 percent. Two separate statistics. They
are both valuable to us.
Ms. Clarke. Yes. There is also the false positive, the cost
of the false positive. That individual that is detained for
whatever reason because there is a false positive, the cost of
that person's health, the cost of that person's well-being,
perhaps there is a commerce concern involved. I am concerned
about the lack of accuracy. I am very concerned about the
lack----
Mr. Wagner. If the person doesn't match the photo in this
case, they present their passport as they are doing today.
Ms. Clarke. Excuse me?
Mr. Wagner. If a person doesn't match a photograph, they
simply present their passport and their boarding pass.
Ms. Clarke. If they trying to match them and they don't
match, what happens to that individual?
Mr. Wagner. They present their boarding pass and their
passport----
Ms. Clarke. Uh-huh.
Mr. Wagner [continuing]. And it is manually reviewed at
that point in time, just as happens today.
Ms. Clarke. Is that--and those people are not detained in
any way? They are not asked to step aside, they are not asked--
the process does not delay that person?
Mr. Wagner. No. They just show their passport.
Ms. Clarke. OK. I hope that is the case.
Will CBP commit to tracking a more meaningful statistic
that captures the usefulness and accuracy of the full facial
recognition process, including the rate at which the system
fails to capture a quality image?
Mr. Wagner. We do track those rates. We track the--what we
call the gallery completion rate. We are never going to have
100 percent of a gallery because not everybody needs a passport
to travel.
Ms. Clarke. Including the images that are not high-quality,
those that fail to meet your standard?
Mr. Wagner. Right. We want to build it so that the camera
will take a high-quality photograph.
Ms. Clarke. I know that is what you want to do. But will
you be keeping statistics on what doesn't meet that standard?
Mr. Wagner. So we are, correct, yes.
Ms. Clarke. Very well.
I yield back, Mr. Chairman.
Chairman Thompson. The Chair recognizes the Ranking Member.
Mr. Rogers. Thank you, Mr. Chairman.
I just want to clarify with the Secret Service: The
information that you have collected in this pilot program that
you talked about earlier, is it my understanding that everybody
that is in that are employees of the Secret Service, and they
volunteered to be in it?
Mr. Di Pietro. That is correct. Maybe if I can explain how
we are doing the pilot, that might help.
Mr. Rogers. Also, when did the pilot start?
Mr. Di Pietro. So we published the PIA back in November, it
began in December, and it is going to run through August. We
did that on purpose. We wanted it to go from the winter into
the summer because of the different items people wear, so that
we have a good amount of time where we were assessing it.
Maybe if I just explain a little bit of how the pilot is
working, that might help explain this for you.
As you indicated, the participants of the pilot are Secret
Service employees who volunteer to take part in this effort.
The facial images are stored when associated match is
recognized on an individual, on one of the volunteers. At the
conclusion of the pilot, all of that information will be
deleted.
We are using our current CCTV system, video management
system we have at the White House. I can imagine you have got a
similar system up here on Capitol Hill that you use for CCTV
surveillance. We are using those video feeds there, and we are
trying to match the individuals that are in the pilot, the
volunteers, to the people who we are seeing in those cameras.
If there is no match, there is no record. If there is a match,
then there is a record. That will be retained till the end of
the pilot, and then that information will be deleted at the
conclusion of the pilot.
Mr. Rogers. Thank you, Mr. Chairman.
Chairman Thompson. Thank you. But I think Mr. Katko's
question was, if you were collecting data, capturing data, and
you said no. My question is, whether it is a volunteer or a
person walking the street, you are collecting data?
Mr. Di Pietro. That is correct. That is right.
Chairman Thompson. The Chair now recognizes the gentleman
from Louisiana, Mr. Higgins.
Mr. Higgins. Thank you, Mr. Chairman.
Director Romine, would you describe biotech--biometric
technology and facial recognition technology as designed to
work with trained agents? In other words, man and machine
working together? Is this what this is working toward?
Mr. Romine. We are agnostic as to whether that is the use
case or not. But our testing has verified that, in the case of
facial recognition, the best algorithms and the best human face
recognizers, the trained face recognizers----
Mr. Higgins. I thank you for pointing that out. In your
testimony, NIST has researched in an effort to measure the
accuracy of forensic examiners, including forensically-trained
facial reviewers.
Mr. Romine. That is correct.
Mr. Higgins. Your statement stated that it presented data
comparing state-of-the-art facial recognition algorithms with
the best human face identifiers, the best machines performed in
the range of the best performing humans----
Mr. Romine. That is correct.
Mr. Higgins [continuing]. Who are professional facial
examiners. But you went on to state that optimal face
identification was achieved only when humans and machines
collaborated.
Is that an accurate assessment?
Mr. Romine. That is correct.
Mr. Higgins. Let me ask, Commissioner Wagner, is there ever
an arrest made or denial to travel based solely on facial
recognition technology?
Mr. Wagner. No.
Mr. Higgins. Thank you.
So facial recognition technology gets a--let's call it a
hit, a high probability based on algorithms, that a particular
traveler is a person of interest. Then an agent looks into the
documentation further and has personal interaction with that
individual, which then approves the individual for travel or
prompts further and deeper investigation. Is that correct?
Mr. Wagner. Yes, that is correct.
Mr. Higgins. So, just to clarify for America watching, this
technology is being used to enhance the efficiency and the
speed by which the trained agents can move travelers through
screening points. Is that contract?
Mr. Wagner. Yes.
Mr. Higgins. Thank you for clarifying that.
Is the general consensus amongst travelers and airlines
that this technology is a good idea, is working well?
Mr. Wagner. I believe so, yes.
Mr. Higgins. Thank you for clarifying that.
Let me jump into your data breach. It is a concern for all
of us, regardless of which side of the aisle we are on.
Who reported that breach? Did they self-report, or was it
discovered? How was it discovered? Is my first two questions
about that. Who reported it, the contractor, or did you all
discover it?
Mr. Wagner. No. I believe we asked them about it.
Mr. Higgins. How much time went by?
Mr. Wagner. A significant amount of time. I need to verify
this, but my recollection seems to be that we asked them if any
of our data was included in it, and they came back and said
yes.
Mr. Higgins. Not to put you on the spot here, my brother,
but I am going to. When you say an amount of time, a pretty
significant amount of time, are you talking days, weeks,
months?
Mr. Wagner. I have that answer. Let me look for that, and I
will come back to you.
Mr. Higgins. OK. We would like to know that, because it
is--the contract was referred to as subsequently terminated. We
would like very much to know what the course of events were
regarding--what was the time line here with this contractor
from the time the breach happened till the time it was
discovered and inquired about and reported and verified, and
then how much time before that contract was terminated?
I believe--I would like to know, and perhaps my colleagues
would like to know, if that contractor is still on the
contracting list? If that contract was terminated with that
contractor, but are they still out there bidding on other
contracts? I believe we would like to know that.
Commissioner Wagner, you have a tremendous job to do.
You, gentlemen, thank you for your service, all of you. It
is important to the Members of this committee to get things
right.
Many ports of entry, particularly land ports, face unique
challenges implementing the biometric entry/exit system.
Can you just share what--this is my final question--what
are the primary challenges and how can we help?
Mr. Wagner. The primary challenge was finding a way to
implement this into a travel system that wasn't designed to
support the, say, collection of biometrics on only a segment of
the traveling public.
You know, unlike Europe and Asia and other places, we don't
have departure controls. You don't see a CBP Officer to get
your passport stamped to depart the United States. We have
never restricted departures like that. So international flights
comingle with domestic flights. Then with each individual
flight, you have got U.S. citizens, permanent residents, and
visitors. So how do you sift and sort and differentiate between
who is in scope or out of scope of the biometric exit
requirement, what technology do you use to collect that
biometric, and how do you ensure a way that is not going to
create gridlock at the airports or the seaports or the land
border, when we get to it, on how to do that.
Mr. Higgins. That is exactly what you are working through
right now, correct?
Mr. Wagner. Right. So we found a way using the facial
recognition and compare people against data they have already
provided in a convenient, quick, and accurate way that we can
apply to all travelers using different authorities and help the
airlines board the planes even faster.
Mr. Higgins. Mr. Chair----
Thank you for that answer.
My time has expired, Mr. Chairman.
Chairman Thompson. Thank you very much.
The Chair recognizes the gentlelady from New Jersey for 5
minutes, Mrs. Watson Coleman.
Mrs. Watson Coleman. Thank you, Mr. Chairman.
Thank you, gentlemen, for your testimony. This is a very,
very important issue for us. I mean, we want to be safe and
secure, but we also want to recognize that our privacy is our
privacy, and we have guarantees under the Constitution and that
we are not in any way infringing upon that.
Mr. Wagner, I would like to ask you a question. I
understand the Department has sent an interim final rule to OMB
that would expand CBP's collection of biometric data, something
we have obviously expressed tremendous interest in.
The committee is eager to learn as much as possible about
what you intend with this rule and why you haven't pursued a
more transparent and deliberative process.
What does this interim final rule entail, how does it
address CBP's collection of biometric data on U.S. citizens,
and why did you choose this closed process rather than
providing notice and allowing public comment?
Mr. Wagner. There are several pieces of rulemakings
underway. There is an interim final rule that is drafted and is
circulating through the Government for comment. There is also
notice of proposed rulemakings on other parts of what we would
like to propose to do. We are evaluating all of those right now
based on a lot of the comments we have received back from
within Government, and we may take a different approach.
There are regulations in place already, though, concerning
biometric exit that have been in place that we are utilizing
today. Through the privacy impact assessments, we have
explained in great detail--in greater detail than would be in
the regulations probably even--how the program operates and
what exactly what happens with it. That is publicly available.
Mrs. Watson Coleman. Are you having conversations with
stakeholders?
Mr. Wagner. Absolutely. I have personally done meetings
with--two different meetings, the East Coast and West Coast,
with the privacy community and all of the privacy
representatives. We are certainly talking with all of our
travel and tourism stakeholders. There is vehement support
behind this in the travel and tourism arena. Of course, we are
talking with the airlines and airports and our Government
partners as well.
Mrs. Watson Coleman. Why is it that I am asking you this
question about why the committee doesn't have the information
it needs? If these discussions have been in the public realm,
why am I asking you about this process? What part of this
process fits this question about why you have chosen to do it
in a more closed way as opposed to a more transparent way? Or
am I just misunderstanding and just misstating?
What part of your consideration, your rulemaking request,
your request to OMB, don't fit this sort of public sharing?
Mr. Wagner. I am not sure I understand the question.
Mrs. Watson Coleman. Well, according to the information
that I was given, the Department has sent an interim final
report to OMB. This interim report has to do with expanding
your collection of biometric data and that the process that you
all are using in dealing with OMB has been a closed process.
What does that mean?
Mr. Wagner. So there are certain provisions that would be
in the interim final rule that, if OMB were to approve it, we
could publish that in the Federal Register. You can still
accept comments, I believe, on that, but the rule goes into
effect.
Really that----
Mrs. Watson Coleman. What is the problem with there being a
more open process now as----
Mr. Wagner. We are doing that, too, for the other
provisions.
Mrs. Watson Coleman. Well, what about the provisions--I am
specifically asking about the provisions that you are not doing
it on. What is the reason for that?
Mr. Wagner. I am----
Mrs. Watson Coleman. Alright. So have you a number of
proposals, rulemaking proposals----
Mr. Wagner. Correct.
Mrs. Watson Coleman. Right?
Part of this, the Department has sent a final--an interim
final rule to OMB. In this particular rule, it deals with the
expansion of CBP's collection of biometric data.
The understanding that I have been given is that the
process that you are engaging in is a closed process, and we
don't have--the committee doesn't have the benefit of what is
being considered, what you are asking for. Instead, you have
used another process that forecloses that opportunity.
So I am asking, why would you choose to do that? What is it
that you are asking for that you can't share in the asking? Not
after the fact.
Mr. Wagner. Well----
Mrs. Watson Coleman. Or is there not such a thing, and we
are just completely uninformed?
Mr. Wagner. No. It is just the different portions of
rulemaking process. Before the rule is even finalized, it would
be premature to talk about what is in it or what is not in it,
because that is going to change. Based on the feedback and our
discussions with OMB, it is going to change.
Mrs. Watson Coleman. But you do that on other rulemaking
requests, but not on this specific area?
Mr. Wagner. We will be publishing a notice of proposed
rulemaking with anything that would fall within those
parameters.
Mrs. Watson Coleman. It is somewhat frustrating----
Chairman Thompson. What I think the point is, at this
point, the public has no input in this process, as far as we
understand.
Mrs. Watson Coleman. Yes.
Chairman Thompson. The rulemaking process.
Normally the notice for rulemaking----
Mr. Wagner. Right.
Chairman Thompson [continuing]. You push it out and receive
comment.
Mr. Wagner. We will do notice of proposed rulemakings to
solicit that feedback.
Chairman Thompson. You will?
Mr. Wagner. We will.
Mrs. Watson Coleman. OK. After that.
Chairman Thompson. We finally got to where we--OK.
Mrs. Watson Coleman. May I just have 30 seconds, since you
so generously----
Chairman Thompson. I will give the lady an additional 30
seconds.
Mrs. Watson Coleman. I am just sort-of curious about the
Secret Service pilot project, and I wanted to understand--I
understand that you are using this pilot project now with
volunteer Service agents, so that when they are walking, you
collect that information, if it matches, it works.
Are you incidentally collecting other information on people
who are not part of this voluntary effort? If so, what are you
doing with those sort of pictures that you capture?
Mr. Di Pietro. So, ma'am, the cameras that we are using as
part of this pilot are part of the White House video management
system. That is the CCTV system that records videos from all of
the cameras around the complex. We retain that data for 30 days
as part of the CCTV process.
If we are--as we are going through and we are identifying
those volunteers that are in there, that record is saved, and
we save that and we are going to evaluate that to the end of
the process.
Mrs. Watson Coleman. But do you have the opportunity to
review other--other faces that you are capturing that are in
the vicinity, tourists, demonstrators, whatever?
Mr. Di Pietro. If it would be something like a false
positive, somebody who wasn't in our pilot but thought it was,
that image would be retained in the----
Mrs. Watson Coleman. We are concerned about what happens
with----
Chairman Thompson. Part of--we will have a Classified
briefing.
Mrs. Watson Coleman. Thank you.
Chairman Thompson. We will have a lot of those questions
responded to.
Mrs. Watson Coleman. Thank you for your extension of time.
Thank you very much for your----
Chairman Thompson. Thank you.
The Chair recognizes the gentlelady from Arizona, Mrs.
Lesko, for 5 minutes.
Mrs. Lesko. Thank you, Mr. Chairman.
First, if you don't mind, I would like to yield a few
seconds to my colleague, Mr. Higgins.
Chairman Thompson. 5 minutes.
Mr. Higgins. Thank you, ma'am.
Mr. Chairman, I ask unanimous consent to enter into the
record 2 op-ed articles in support of law enforcement
application of biometric technology.
The first is from New York City Police Commissioner James
O'Neill, and the second is from managing director of the
Chertoff Group, Lee Kair.
Chairman Thompson. Without objection.
[The information follows:]
How Facial Recognition Makes You Safer
Used properly, the software effectively identifies crime suspects
without violating rights.
By James O'Neill, June 9, 2019, New York Times.
In 1983, when I was sworn in as a police officer, many of the
routine tasks of the trade would have seemed more familiar to a cop
from my grandfather's day than to a new police academy graduate today.
I took ink fingerprints on paper cards and used a Polaroid camera for
mug shots. Reports were handwritten or typed on carbon triplicates.
Biological evidence could be analyzed only in terms of blood type.
Technology has improved the profession beyond what the most
imaginative officer could have conceived in those days. These
innovations include facial recognition software, which has proved its
worth as a crime-fighting resource since we adopted it in 2011. But the
technology has also raised concerns about privacy, so the public should
know how the New York Police Department uses its system--and the
safeguards we have in place.
When detectives obtain useful video in an investigation, they can
provide it to the Facial Identification Section, of the Detective
Bureau. An algorithm makes a template of the face, measuring the shapes
of features and their relative distances from each other. A database
consisting solely of arrest photos is then searched as the sole source
of potential candidates--not photos from the Department of Motor
Vehicles, Facebook, traffic cameras or the myriad streams of closed-
circuit TV video from around the city. Facial ``landmarks'' are
compared without reference to race, gender or ethnicity.
After the software generates a list of possible matches, an
investigator assesses their resemblance to the suspect. If one is
selected, a review is conducted by detectives and seasoned supervisors,
noting similarities and differences. If they affirm the match, the
investigator proceeds with further research, including an examination
of social media and other open-source images.
We might find social media images of a person at a birthday party
wearing the same clothing as the suspect in a robbery. That person then
becomes a lead; the facial identification team will provide only a
single such lead to the case detective. Leads provided by the unit are
comparable to tips to our Crime Stoppers hotline--no matter how
compelling, they must be verified to establish probable cause for an
arrest. No one can be arrested on the basis of the computer match
alone.
In 2018, detectives made 7,024 requests to the Facial
Identification Section, and in 1,851 cases possible matches were
returned, leading to 998 arrests. Some investigations are still being
conducted and some suspects have not been apprehended.
But in many cases there have been clear results. Recently, the work
of the facial identification team led to the arrest of a man accused of
raping a worker at a day spa, and another charged with pushing a subway
passenger onto the tracks. We have made arrests in murders, robberies
and the on-air assault of a TV reporter. A woman whose dismembered body
was found in trash bags in two Bronx parks was identified. So was a
woman hospitalized with Alzheimer's, through an old arrest photo for
driving without a license.
The software has also cleared suspects. According to the Innocence
Project, 71 percent of its documented instances of false convictions
are the result of mistaken witness identifications. When facial
recognition technology is used as a limited and preliminary step in an
investigation--the way our department uses it--these miscarriages of
justice are less likely.
We have never put police sketches into the system; they would be of
no value. We have used editing software to substitute a generic feature
when a suspect is closing his eyes or sticking out his tongue in the
submitted photo. The system can also create a mirror image of the right
side of a face if we have only the left side, for example, to produce a
3-D model.
We use these methods solely to fill in missing or distorted data.
And when we do so, we bring an additional degree of scrutiny to the
process. To compare this to filling in a partial fingerprint, as the
Georgetown Center for Privacy and Technology did in a recent report, is
absurd. It makes sense to create an image of a suspect's left ear using
his right ear as a model. But it is impossible to infer the shape of a
nose from the shape of a chin. As the algorithm is constantly improving
in its ability to read lower-quality images, the editing software is
used less and less frequently.
The department does not conduct civil immigration enforcement, and
neither does our Facial Identification Section. But we do work with
other police departments when appropriate. A recent request from the
F.B.I. led to the identification of a child sex trafficker who
advertised his services on social media.
Biometric technology is no longer new. It is routinely used
everywhere from shopping malls to doctors' offices. Its application by
the department is carefully controlled and its invaluable contributions
to police investigations have been achieved without infringement on the
public's right to privacy. When cases using this technology have been
prosecuted, our methods and findings are subject to examination in
court.
Facial recognition technology can provide a uniquely powerful tool
in our most challenging investigations: When a stranger suddenly
commits a violent act on the street. In the days of fingerprint cards
and Polaroid mug shots, these crimes defined New York City, for
visitors and residents alike.
Though far rarer now, they remain life-altering, and sometimes
life-ending, events. To keep New York City safe requires enormous and
relentless effort. It would be an injustice to the people we serve if
we policed our 21st-Century city without using 21st-Century technology.
James O'Neill is the police commissioner for New York City.
______
Biometrics can protect our borders--along with our privacy
By Lee Kair, opinion contributor, 05/09/19 11 o'clock AM EDT,
TheHill.com
Customs and Border Protection (CBP) has been expanding its
biometric programs with the use of facial recognition technology for
inbound passengers, achieving early success both in identifying
imposters attempting to enter the U.S. and improving the efficiency of
the screening process itself.
Based on this success, the Department of Homeland Security (DHS)
recently announced efforts to expand programs to those departing the
U.S., with the goal of covering 97 percent of outbound international
travelers in the next 4 years.
As DHS applies facial recognition and other biometric technologies
to confirm travelers' identities and to intercept potential threats, it
is important to look at how it balances travelers' privacy with
security goals.
Not surprisingly, the expanded use of biometrics raises questions
about individual privacy, particularly in light of proliferating, high-
profile data breaches that can affect--and should alarm--all of us.
As the lead agency for protecting our nation's borders, CBP has
evolved its process for identifying and screening passengers over time.
In ``the old days,'' passengers flying into the United States would
present their passport to a CBP officer. The officer compared the
laminated picture within the passport to the person standing in front
of them, researched available government data sources to determine if
the traveler was high-risk, and conducted in-person interviews to
determine if additional screening was necessary. Although a sufficient
process, it was time-consuming and dependent on CBP personnel to make
accurate assessments and detect anomalies in real time.
Since 2005, CBP has required airlines to provide manifest data
shortly after departure so officers can leverage existing targeting
infrastructure and resources, including government documentation and
photographs (such as passport and visa photos), to determine the risk
of incoming passengers before they arrive. Upon landing, low-risk
passengers are expedited through customs while CBP focuses its
resources on higher risk passengers.
Today, CBP is leveraging commercially available biometric
technologies to streamline and automate the existing process of
manually matching images from data bases to individual travelers
attempting entry into or exit from the U.S. The Transportation Security
Administration (TSA) and aviation industry partners also are conducting
biometric pilots across the country to expedite the traveler experience
at the airport. These pilots are intended to confirm the identity of
traveling passengers at various points in the airport ecosystem, with
the goal of enhancing security while reducing friction in the travel
process.
As stakeholders evaluate CBP's deployment of biometric technology,
there are three areas where CBP has demonstrated best practices that
meet the goal of promoting both security and an improved traveler
experience. These include leveraging new technology for more efficient
and effective screening; providing transparency around the collection
and use of biometrics in the screening process; and voluntary opt-in or
opt-out participation for other biometric programs:
Transparency.--CBP and TSA have issued several Systems of Records
Notices and Privacy Impact Assessments while inviting public comment
and publicizing strategies and roadmaps to educate and inform
stakeholders on the steps they are taking to leverage technology for
the security of the traveling public. This level of transparency is
critical to developing trust between travelers and the government. In
an era in which commercial companies often use ``terms of service''
obfuscated with pages of legal language, the government is being clear
about its use of biometrics.
Leveraging existing systems to make them more efficient.--Where the
government already had access to--and used--biometrics through existing
systems (such as photos from passports, visas, previous border
crossings or trusted-traveler programs), the use of matching technology
expedites old manual processes. This speeds the traveler experience and
is more effective than manual visual comparisons. For example,
automated matching of a facial or fingerprint biometric at the TSA
screening checkpoint is likely more accurate and faster than a security
officer's visual driver-license check. These enhancements allow TSA to
increase speed and security while reallocating officer resources to
focus on detecting additional threats to aviation security.
Voluntary use.--CBP and TSA strategies also require the ability to
opt in or opt out of other biometric matching programs and third-party
use of biometrics. Specifically, CBP programs allow passengers to opt
out of technical demonstrations as well as the sharing of biometric
information with third parties (such as airlines); TSA requires opt-in
participation for its biometric trusted traveler programs at TSA
checkpoints.
Many privacy advocates are concerned that the government could use
the data for continuous surveillance without any suspicion of
wrongdoing, to identify and track people without their knowledge.
Critics claim that it's an overreach for the government to require U.S.
citizens to submit to facial scans to board a plane.
However, it is important to point out that CBP privacy policies
only allow the biometric data to be used for identification purposes
and that it must be deleted within 12 hours, in the case of U.S.
citizens. Similarly, TSA is limiting its biometric programs to trusted-
traveler programs, in which travelers have already chosen to share
information.
In a time when we have seen rising concerns about stockpiling user
data on social media, the use of biometrics by both the government and
commercial entities must continue to be evaluated. Countries around the
world are assessing the privacy exposure related to biometrics and
facial recognition. The potential for commercial entities to combine
biometric data with other user data--including geolocation, online
activity and retail purchases--has the potential to significantly
expose sensitive information about private citizens.
While DHS's pilot programs must be evaluated on a continuous basis,
I believe that DHS has handled the implementation correctly. This
should be the standard for other organizations and government entities
looking to deploy biometric-based solutions that create a more secure,
trusted environment for the public.
Lee Kair is managing director of The Chertoff Group, a security and
risk management advisory firm. He served more than 15 years in senior
executive positions at the U.S. Department of Homeland Security,
including the Transportation Security Administration. The Chertoff
Group is a frequent adviser to clients in the defense technology and
aviation industries, including clients that work in identity management
and biometrics technology. Follow on Twitter @ChertoffGroup.
Chairman Thompson. The gentlelady is recognized for the
additional time.
Mrs. Lesko. Thank you, Mr. Chairman.
I, too, ask unanimous consent to enter into the record 3
letters expressing support for the effective and responsible
use of biometrics by TSA and CBP. These letters are from
Airlines for America, the International Air Transport
Association, and the Global Business Travel Association.
Chairman Thompson. Without objection.
[The information follows:]
Letter From Sharon Pinkerton to Chairman Bennie G. Thompson and Ranking
Member Mike Rogers
July 8, 2019.
The Honorable Bennie Thompson,
Chairman, Committee on Homeland Security, U.S. House of
Representatives, 2466 Rayburn House Office Building,
Washington, DC 20515.
The Honorable Mike Rogers,
Ranking Member, Committee on Homeland Security, U.S. House of
Representatives, 2184 Rayburn House Office Building,
Washington, DC 20515.
Dear Chairman Thompson and Ranking Member Rogers: Over the past
decade, the Department of Homeland Security (DHS) has been evaluating
and testing approaches to determine the most effective manner to add
biometrics to its arrival and departure procedures to provide better
security while maintaining privacy and facilitating the travel
experience. We support those efforts.
The work is being done so that the U.S. Customs and Border
Protection (CBP) can implement the congressional mandate to administer
a biometric air entry/exit program for departing international air
passengers. The Transportation Security Administration (TSA) also is
evaluating biometrics for identity verification at the security
checkpoint. The primary benefits of the biometric programs are the
enhanced ability to protect against identification fraud and to improve
DHS's ability to determine the rate of visa overstays.
A4A members have worked closely with CBP and TSA during this
process and participated in the DHS Science and Technology
Directorate's technology evaluations and pilot programs. DHS has worked
to address and meet our principal goals of ensuring that any biometric
program would increase security, improve the passenger experience and
not require airlines to perform government functions.
The security benefits of biometrics are undeniable. For example,
the CBP match rate associated with facial recognition technology is
consistently high, above 98 percent, and it is expected that technology
will continue to improve. TSA, through collaboration with CBP, also is
seeing the benefits of biometric technology, in particular facial
recognition technology.
While we believe the privacy protections currently in place are
effective, we will continue to work with the DHS, CBP, TSA and our
passengers to ensure the highest levels of privacy. Airlines already
collect and transmit biographic data to DHS to comply with Federal
security requirements, so we have experience in the area. Airlines,
like DHS, also have committed to strict privacy principles as it
relates to the use of biometric information. For facial recognition
technology, these principles include opt-out options and non-retention
of photos for business purposes. In fact, airlines and airports must
immediately purge images following transmittal to CBP for identity
verification. We all agree that privacy is of the utmost importance.
We appreciate the collaboration that DHS has demonstrated in
implementing the statutory mandate to administer biometrics to improve
our nation's security. We recognize this is an area of rapidly changing
technology and public acceptance and we look forward to working with
Congress and the Administration to continue to make our nation's
aviation system even more secure while improving the passenger
experience. We believe that Congress can play a constructive role in
incentivizing the best biometrics technology and we look forward to
working with you as the technological capabilities continue to advance.
Sincerely,
Sharon Pinkerton
Senior Vice President, Legislative & Regulatory Policy, Airlines
for America.
______
Letter From Douglas E. Lavin to Chairman Bennie Thompson and Ranking
Member Mike Rogers
July 8, 2019.
The Honorable Bennie Thompson,
Chairman, Committee on Homeland Security, U.S. House of
Representatives, Washington, DC 20515.
The Honorable Mike Rogers,
Ranking Member, Committee on Homeland Security, U.S. House of
Representatives, Washington, DC 20515.
Dear Chairman Thompson and Ranking Member Rogers: On behalf of the
International Air Transport Association (IATA) and its 290 member
airlines, I appreciate the opportunity to comment on the use of
biometric technologies in aviation. IATA is a strong supporter of the
use of biometrics to facilitate a safe, secure, and efficient
commercial air travel experience for our members' passengers.
IATA estimates that the number of airline passengers globally will
double by 2037. Given that aviation infrastructure development (e.g.
airports and air traffic management) will likely not be able to keep
pace with such growth, IATA has undertaken several initiatives designed
to improve the experience and efficiency of the current travel process,
particularly passenger facilitation at airports. IATA's ``One ID''
program seeks to introduce a streamlined, friction-free, and passenger-
centric process that allows an individual to assert their identity to
the required level at every process step while maintaining the privacy
of personal data. One ID is premised on a single token biometric that
can be used at each touchpoint across the end-to-end journey.
In June 2019, the IATA Annual General Meeting passed the attached
resolution on the One ID program which affirms the significant benefits
of paperless travel by means of biometric recognition and encourages
governments to collaborate on a biometric-based identify management
solution.
IATA and several our member airlines have also worked closely with
U.S. Customs and Border Protection (CBP) on their proposed biometric
entry/exit system. We are very pleased with CBP's engagement with
industry on this program and their consideration of important
operational issues and the protection of passenger privacy. We look
forward to continuing to work with CBP in a collaborative fashion as
they implement this system.
Thank you for your consideration.
Sincerely,
Douglas E. Lavin,
Vice President, Member and External Relations--North America.
Attachment.--RESOLUTION ON ONE ID
RECALLING that global air passenger traffic is projected to double
by 2037, meaning that the air transport sector will accommodate an
additional four billion passengers by this time;
FURTHER RECALLING that the practical obligation to obtain and check
passenger identity documentation and travel authorizations is often
placed upon carriers as a part of immigration and border security
processes;
ACKNOWLEDGING that a safe, secure and seamless passenger experience
is an objective of primary importance for consumers, governments and
the airline industry;
RECOGNIZING that efficient and optimized communication standards
support both enhanced customer experiences and more effective security
outcomes;
FURTHER RECOGNIZING that important shifts in consumer behaviour,
together with changing expectations in respect of real-time
information, paperless processes and data privacy, require a high
degree of collaboration between air transport sector stakeholders;
The 75th IATA Annual General Meeting:
1. AFFIRMS the significant benefits of paperless passenger travel
by means of biometric recognition;
2. ENCOURAGES government authorities, member airlines and airports
to support the One ID strategy;
3. ENCOURAGES ICAO and its member states to urgently identify
specifications for a digital travel credential that will offer
a secure and efficient alternative to passports;
4. ENCOURAGES member airlines and all other actors in the air
transport system to work together toward a ``use case'' for
such a globally accepted digital travel credential;
5. CALLS on government authorities, member airlines and airports to
urgently:
(i) collaborate on identity management solutions for the
sharing of identity information to avoid duplication in
passenger checks and enable secure paperless processes, with
such solutions to satisfy the highest security principles and
meet the important requirements of privacy law;
(ii) work together to find interoperable and innovative
solutions;
(iii) further explore and apply the benefits of biometric
recognition, including in terms of security and speed;
6. ENCOURAGES governments to explore the possibility of offering
the verification of passenger identity information as a
service.
______
Letter From Shane C. Downey to Ranking Member Mike Rogers
July 9, 2019.
Ranking Member Mike Rogers,
House Homeland Security Committee, H2-117 Ford House Office Building,
Washington, DC 20515.
Dear Ranking Member Rogers: The Global Business Travel Association
(``GBTA'') is the world's premier business travel and meetings trade
organization headquartered in the Washington, DC area with 40 State
chapters in the U.S. and operations on six continents. GBTA's 9,000-
plus members manage more than $345 billion of global business travel
and meetings expenditures annually. GBTA delivers world-class
education, events, research, advocacy, and media to a growing global
network of more than 28,000 travel professionals and 125,000 active
contacts.
GBTA members work for the majority of Fortune 100 companies,
buying, sourcing and managing the corporation's travel budget, among
other responsibilities like ensuring the safety and security of their
travelers. In a well-managed travel program, a corporation can see a
return on investment of $20 for every $1 spent.
Air travel is a major part of business travel and corporate spend.
GBTA research on the U.S. economic impact of business travel shows 515
million domestic business trips are taken in a year. Nearly 30 percent
involve air travel meaning business travelers take to the skies on over
144 million trips a year.
Because of this mass of travelers, GBTA has made secure and
efficient travel a key platform of GBTA's legislative policy. GBTA has
consistently called for increased security at airports, including the
hiring of additional officers to man these critical areas. However, as
travel in general, and business travel specifically, has continued to
grow exponentially, it has become clear that simply hiring more people
is not enough, and that technology and pre-screening of passengers are
necessary to support a system that is safe and efficient.
GBTA has been a supporter of U.S. Customs and Border Protection's
(CBP) Global Entry and Transportation Security Administration's (TSA)
PreCheck since their inception. GBTA believes the use of biometrics and
facial recognition is the logical next step to further increase
traveler safety and efficiency in moving through security checkpoints.
This support stems from understanding the issues that impact
business travel. GBTA surveys of business travelers consistently cite
moving through airport security as one of the largest pain points.
PreCheck and Global Entry have delivered business travelers a risk-
based, intelligence-driven aviation security system that is safe, fast
and efficient. Time is money for business travelers, and inefficient
procedures reduce business travel due to the ``hassle factor'' and hurt
the economy.
To further illustrate the impact efficient screening can have, look
to GBTA's ``Business Traveler Sentiment Index,'' which profiles
business travelers' attitudes around business travel and how that
impacts their actual travel behavior. Our research shows TSA PreCheck
enrollees are significantly more satisfied with air travel than those
not enrolled. Two-thirds (66 percent) of travelers enrolled in TSA
PreCheck are satisfied with getting through airport security, compared
with just 47 percent of business travelers not enrolled in the program.
More striking is the impact the program has on the overall travel
experience, 66 percent report satisfaction, compared to 54 percent for
those not enrolled.
Today's airport experience involves heavy friction and endless
queuing at the counter check-in, bag drop, security screening and
boarding. As facial recognition security programs expand, meeting the
goal of frictionless travel improves. GBTA supports industry,
governments and travelers working together to create a multi-layered
approach that includes facial recognition for travel security screening
purposes. GBTA believes the business traveling public will continue to
embrace this security tool provided the following continue to progress:
Data security is paramount, and the operators must ensure all
protocols and procedures are followed to ensure the safety of the
individual's data;
False Identification Mitigation must continue to advance and be a
part of all future plans. Prior to the enactment of Secure Flight,
business travelers all too often had their identities confused with
others on flight watch lists, causing delays and unnecessary hassles at
the airport. Without a mitigation strategy in place, the same could
occur with biometrics and facial recognition;
And, travelers are made aware of the ability to opt in and out of
facial screening checkpoints.
GBTA encourages Congress to continue to work with the Department of
Homeland Security and other key agencies, the security industry and
travelers to strengthen and streamline travelers' safety and security.
Sincerely,
Shane C. Downey,
Vice President, Government Relations, Global Business Travel
Association.
Mrs. Lesko. Thank you.
All my questions, Mr. Chairman and Members, are for Mr.
Gould.
My first question, Mr. Gould, is the pilot program that you
have working with Delta down in Atlanta, where do you get the
photos from? Is it opt-in? Do you share--get the database of
passports from CBP?
That is my first question.
Mr. Gould. Yes, ma'am. We use CBP's TVS matching service
for that. CBP has access to State Department photos for the
back-end matching.
Then it is an opt-in program. Passengers have the
opportunity to choose whether to present biometric
identification using the facial capture or to present a
credential. We see very high rate of people choosing to provide
the facial image.
Mrs. Lesko. OK. So just so that I understand, where do you
ask them if they want their photo taken?
Mr. Gould. Ma'am, there are signs throughout the checkpoint
area that say we are piloting this technology and that should
you choose not to participate, please let the TSO, the officer,
know.
As you approach the TDC, the travel document checker
position, there is an officer there. The officer will say, you
know, do you choose to provide biometric identification? In
which case, if the passenger says yes, they are directed to
stand in a specific location for that facial capture. So there
is interaction with the officer at that point.
Mrs. Lesko. Thank you. That is very informative.
My next question is due to, I guess, the success of CBP's
use of biometrics. I think this is--you know, this technology
is going to happen. I do agree with other Members that we need
to make sure that we have privacy and security in it, of
course.
But are you going to use any of the--is TSA planning on
looking at how they can work, I guess, with CBP on their
success in order to implement it in more airports?
Mr. Gould. Yes, absolutely, ma'am. That is the reason we
are doing the pilot in Atlanta, is to understand that
interaction between us and the CBP TVS system and what benefit
that system brings to the TSA checkpoint and the identification
verification process.
Mrs. Lesko. Good. I am glad that you are working on it, and
hopefully we can get a fairly fast turnaround. I probably would
be interested in going and seeing what you are doing down in
Atlanta myself.
Mr. Gould. Yes, ma'am.
Mrs. Lesko. Also, Mr. Gould, are you planning on using
this, or have you thought of using biometric technology, or do
you, for the employees, the airport employees?
Mr. Gould. Yes, ma'am. We are considering using biometric
identification processes for employees as well.
Mrs. Lesko. Thank you.
The reason that I ask that is because from some of our
briefings, hearings, I think we have been concerned about
insider-type threats. I think what happened up in--what was it,
Washington airport? I can't remember where--an employee take a
plane and flew it----
Mr. Gould. Yes, ma'am, Seattle.
Ms. Lesko. Yes, Seattle, Washington. So--and with baggage
handlers and those types of things. So it seems to me that it
would be logical that we use biometric screening for the
employees themselves.
Mr. Gould. Yes, ma'am. That is certainly something we will
be looking at.
Mrs. Lesko. Thank you.
I yield back my time.
Mr. Payne [presiding]. Thank you. We now recognize the
gentlelady from Texas for 5 minutes of questioning.
Ms. Jackson Lee. Mr. Chairman, thank you very much.
I want to start off by asking unanimous consent to put into
the record an op-ed by the Houston Chronicle, Real Abuses at
the Border: Squalid conditions for detained migrants are worthy
of all outrage Americans can muster. Ask unanimous consent.
Mr. Payne. Without objection.
[The information follows:]
Border Patrol abuses real, and worthy of outrage
The Editorial Board, July 5, 2019, Updated: July 5, 2019 8:59 a.m.,
Houston Chronicle
A ticking time bomb.
That's how a senior manager described the situation at a Border
Patrol detention facility in the Rio Grande Valley, according to a
report by the Office of Inspector General released this week. The
independent watchdog's findings describe squalid, overcrowded
conditions at several facilities, where men, women and children are
poorly fed and held without access to showers, sometimes for weeks.
The investigators' words and images--men crammed together in
standing-room-only cells, dozens of women and children lying side to
side on concrete floors--support the testimony of doctors and lawyers
who spoke out last week after interviewing immigrants in detention.
They also lend credence to the stories Democratic lawmakers heard
during a recent visit to a holding facility outside El Paso.
Some had dismissed these claims as politically self-serving, or as
the embellishments of partisans and activists looking to gin up
outrage. Turns out the government's own reporting shows conditions at
these detention centers are worthy of all the outrage Americans can
muster.
Along with overcrowding, investigators found more than 800 of the
2,669 children in custody at the facilities had been held longer than
72 hours, violating a court settlement as well as Customs and Border
Protection policy. This included a group of 50 unaccompanied children
under 7 years old, some of whom had been in these deplorable
circumstances for more than 2 weeks.
This report follows a similar warning by the OIG in May after a
visit by investigators to holding locations in the El Paso area.
The excuse that the government has been overwhelmed by the number
of arrivals, many asylum-seekers from Central America, has worn thin.
During a previous increase under the Obama Administration in 2014,
mostly by unaccompanied minors, officials were also unprepared. Yet
they quickly opened detention space across the country while officials
made arrangements for the children to be released as quickly as
possible into the custody of family or other sponsors. Although it was
far from an ideal situation--this is where the first images of
``children in cages'' came from--it relieved overcrowding and sped up
processing time.
The numbers this time around are larger, but the response has been
anemic--seemingly, by design.
So far, the lack of urgency in easing these conditions fits
squarely into the spirit of deterrence through pain that has been at
the heart of U.S. immigration policies over the years, but which have
hardened unconscionably during the Trump administration.
The need to quickly move detained immigrants from Border Patrol
custody intensified on Monday, after some of the exchanges of a private
Facebook group were released. As reported by ProPublica, group members,
including current and former Border Patrol agents, posted racist,
sexist and violent memes about immigrants and New York Congresswoman
Alexandria Ocasio-Cortez.
On an image of a migrant fording the Rio Grande while dragging a
young boy in a plastic bag, group members wrote disparaging comments
such as, ``At least it's already in a trash bag.'' Under a photo of a
father and his 23-month-old daughter who drowned in the river, the
member who posted the image asked if it was fake because the
``floaters'' were so ``clean.''
The revelation of the Facebook group comes on the heels of text
messages between Border Patrol agents made public as part of an ongoing
court case in Arizona, where an agent is accused of knocking down a
Guatemalan man with his vehicle and then covering it up. In one
exchange, the agent refers to immigrants as ``disgusting subhuman s---
unworthy of being kindling for a fire.''
All these statements are vile and intolerable, but this isn't just
name calling. When these attitudes are brought to bear, they can mean
the difference between life and death. Between ignoring the jugs of
water that humanitarian groups leave for migrants in the desert or
slashing and stomping them. Between taking cover with your fellow
agents as rocks fly overhead from across the border fence or
indiscriminately shooting into Mexico at anything that moves.
Of course, that side of Border Patrol is countered with the many
agents who act humanely while fulfilling their duties, who put their
lives on the line to protect immigrants and enforce our laws. But even
some of the good actors are pressured to remain silent by a culture
that protects its own, no matter the cost, while whistleblowers are
ostracized.
Tolerance of these attitudes has gone on long enough.
The House Judiciary and Oversight Committees announced hearings
next week into the conditions at detention centers. That's a good
start. The agents who violated policy, and basic human decency, should
also be punished. And over time, leadership should not only set an
example, but work to improve the culture at the Border Patrol, which
for far too long has gotten away with little accountability or
transparency.
Meanwhile, the time bomb keeps ticking.
Ms. Jackson Lee. Ask unanimous consent from the CNET
article, Monday, July 9, Acting DHS Secretary Defends Border
Conditions.
Mr. Payne. Without objection.
[The information follows:]
DHS Official Defends Conditions at Border Patrol Stations
July 8, 2019.
WASHINGTON (AP).--Acting Homeland Security Secretary Kevin
McAleenan on Sunday defended conditions at U.S. Border Patrol stations
following reports of crowded and unsanitary conditions that have
heightened debate about President Donald Trump's immigration policy, a
trademark issue for his reelection campaign.
``It's an extraordinarily challenging situation,'' McAleenan told
ABC's ``This Week.''
The Homeland Security Department's internal watchdog provided new
details Tuesday about the overcrowding in Texas' Rio Grande Valley, the
busiest corridor for illegal crossings. The report said children at
three facilities had no access to showers and that some children under
age 7 had been held in jammed centers for more than 2 weeks. Some cells
were so cramped that adults were forced to stand for days on end.
Government inspectors described an increasingly dangerous
situation, both for migrants and agents--a ``ticking time bomb,'' in
the words of one facility manager. The report echoed findings in May by
the department's inspector general about holding centers in El Paso,
Texas: 900 people crammed into a cell with a maximum capacity of 125;
detainees standing on toilets to have room to breathe; others wearing
soiled clothing for days or weeks.
In tweets Sunday afternoon, Trump went further than McAleenan in
defending his administration's response, accusing the news media of
``phony and exaggerated accounts'' but without providing evidence.
``Border Patrol, and others in Law Enforcement, have been doing a
great job. We said there was a Crisis--the Fake News & the Dems said it
was `manufactured,' '' Trump wrote. Federal detention centers ``are
crowded (which we . . . brought up, not them) because the Dems won't
change the Loopholes and Asylum. Big Media Con Job!''
Democrats faulted Trump for not offering an immigration overhaul
that could pass a divided Congress.
``The president is acting like we are some weak, pathetic
country,'' said Colorado Sen. Michael Bennet, a Democratic Presidential
candidate. ``We have the ability to treat human beings humanely. We
have the ability to lead our hemisphere in a discussion about how to
deal with this refugee crisis,'' he said on ``Fox News Sunday.''
McAleenan said that since the first of the year, 200 medical
providers have been added to facilities, including personnel from the
U.S. Coast Guard and the Public Health Service Commission Corps.
``We have pediatricians in border patrol stations for the first
time in history trying to help address conditions where children are
coming across 300 a day in . . . April and May,'' McAleenan said.
``We've built soft-sided temporary facilities. These are spaces
that are much more appropriate--high ceilings, more room for children
and families. We've put them both in Donna, Texas, in South Texas as
well as in El Paso to provide additional space . . . We've bought buses
to transport people to better places.''
McAleenan disputed news reports, including those by The Associated
Press, of especially troubling conditions at a border station in Clint,
Texas, where a stench was coming from children's clothing and some
detainees were suffering from scabies and chickenpox.
``There's adequate food and water,'' he said. ``The facility's
cleaned every day, because I know what our standards are and I know
they're been followed because we have tremendous levels of oversight.
Five levels of oversight.
``Inadequate food, inadequate water and unclean cells. None of
those have been substantiated.''
He said everyone in the chain of command is worried about the
situation of children detained at the border. He said that on June 1,
his department had 2,500 children in custody, including 1,200 who had
been there for more than 3 days. As of Saturday, McAleenan said there
were 350 children, and only 20 have been in the department's custody
for more than 3 days.
``So that's huge improvement based on the resources we asked for
from Congress and were finally given,'' he said.
Sen. Jeff Merkley, D-Ore., told NBC's ``Meet the Press'' that he is
stunned when administration officials say that reports on the
conditions are unsubstantiated.
``I'm just like, `What world are they living in?' '' Merkley said,
citing government and news reports. ``From every direction you see that
the children are being treated in a horrific manner. And there's an
underlying philosophy that it's OK to treat refugees in this fashion.
And that's really the rot at the core of the administration's policy.''
Separately, McAleenan addressed questions about U.S. Border Patrol
agents who are under fire for posting offensive messages in a
``secret'' Facebook group that included sexually explicit posts about
U.S. Rep. Alexandria Ocasio-Cortez and dismissive references to the
deaths of migrants in U.S. custody. The existence of that group was
reported Monday by ProPublica. Prior to that, few people outside the
group had ever heard of it.
He said an allegation about such activity was investigated in 2016.
``Discipline was meted out on an agent that made an offensive post on
that website,'' he said.
Ms. Jackson Lee. I ask unanimous consent to put into the
record the IG inspector's report, dated July 2, 2019.
Mr. Payne. Without objection.*
---------------------------------------------------------------------------
* The information has been retained in committee files and is
available at https://www.oig.dhs.gov/sites/default/files/assets/2019-
07/OIG-19-51-Jul19_.pdf.
---------------------------------------------------------------------------
Ms. Jackson Lee. I ask to put into the record an article
found in The New York Times, ICE uses facial recognition to
mine State driver's license. That is July 7. Ask unanimous
consent.
Mr. Payne. Without objection.
[The information follows:]
ICE Used Facial Recognition to Mine State Driver's License Databases
By Catie Edmondson, July 7, 2019.
WASHINGTON.--Immigration and Customs Enforcement officials have
mined state driver's license databases using facial recognition
technology, analyzing millions of motorists' photos without their
knowledge.
In at least three states that offer driver's licenses to
undocumented immigrants, ICE officials have requested to comb through
state repositories of license photos, according to newly released
documents. At least two of those states, Utah and Vermont, complied,
searching their photos for matches, those records show.
In the third state, Washington, agents authorized administrative
subpoenas of the Department of Licensing to conduct a facial
recognition scan of all photos of license applicants, though it was
unclear whether the state carried out the searches. In Vermont, agents
only had to file a paper request that was later approved by Department
of Motor Vehicles employees.
The documents, obtained through public records requests by
Georgetown Law's Center on Privacy and Technology and first reported on
by The Washington Post, mark the first known instance of ICE using
facial recognition technology to scan state driver's license databases,
including photos of legal residents and citizens.
Privacy experts like Harrison Rudolph, an associate at the center,
which released the documents to The New York Times, said the records
painted a new picture of a practice that should be shut down.
``This is a scandal,'' Mr. Rudolph said. ``States have never passed
laws authorizing ICE to dive into driver's license databases using
facial recognition to look for folks.''
He continued: ``These states have never told undocumented people
that when they apply for a driver's license they are also turning over
their face to ICE. That is a huge bait and switch.''
The use of facial recognition technology by law enforcement is far
from new or rare. Over two dozen states allow law enforcement officials
to request such searches against their databases of driver's licenses,
a practice that has drawn criticism from lawmakers and advocates who
say that running facial recognition searches against millions of photos
of unwitting, law-abiding citizens is a major privacy violation.
The F.B.I., for example, has tapped state law enforcement's troves
of photos--primarily those for driver's licenses and visa
applications--for nearly a decade, according to a Government
Accountability Office report. The bureau has run over 390,000 searches
through databases that collectively hold over 640 million photos,
F.B.I. officials said.
The Georgetown researchers' documents covered 2014 to 2017, and it
was not immediately clear if those states still comply with the ICE
requests. Representatives for the states' motor vehicles departments
could not immediately be reached for comment Sunday night.
On Monday, Amy Tatko, the public outreach manager for the Vermont
Agency of Transportation, said in a statement that the use of facial
recognition technology by the agency was discontinued in 2017 ``at the
direction of current Governor Phil Scott as soon as it was brought to
his attention.''
Matt Bourke, an ICE spokesman, said the agency would not comment on
``investigative techniques, tactics or tools'' because of ``law-
enforcement sensitivities.''
But he added: ``During the course of an investigation, ICE has the
ability to collaborate with external local, Federal and international
agencies to obtain information that may assist in case completion and
subsequent prosecution. This is an established procedure that is
consistent with other law enforcement agencies.''
The researchers sent public records requests to each state,
searching for documents related to law enforcement's relationship with
state motor vehicles departments. They received varying degrees of
responsiveness but discovered the ICE requests in Utah, Washington and
Vermont, which have come under fire before for sharing driver's license
information with the agency.
The Seattle Times reported last year that Washington State's
Department of Licensing turned over undocumented immigrants' driver's
license applications to ICE officials, a practice its Governor, Jay
Inslee, pledged to stop. And a lawsuit in Vermont filed by an activist
group cited documents obtained under public records law that showed
that the State Department of Motor Vehicles forwarded names, photos,
car registrations and other information on migrant workers to ICE,
Vermont Public Radio reported this year.
The relationship between Washington's Department of Licensing and
ICE officials may prove to be particularly interesting to privacy
experts because of a law the State Legislature passed in 2012
stipulating that the department could use a facial recognition matching
system for driver's licenses only when authorized by a court order,
something ICE did not provide.
Facial recognition technology has faced criticism from experts who
point to studies that show that recognition algorithms are more likely
to misidentify people of color--and in particular, women of color. At
least 25 prominent artificial-intelligence researchers, including
experts at Google, Facebook and Microsoft, signed a letter in April
calling on Amazon to stop selling its facial recognition technology to
law enforcement agencies because it is biased against women and racial
minorities.
The use of the technology has also come under fire from a
bipartisan group of lawmakers. The House Homeland Security Committee,
led by Representative Bennie G. Thompson, Democrat of Mississippi, will
hold a hearing on Wednesday grilling Department of Homeland Security
officials about their use of facial recognition. The chairman of the
House Committee on Oversight and Reform, Representative Elijah E.
Cummings of Maryland, has pledged to investigate the use of the rapidly
expanding technology in the public and private sectors.
``This technology is evolving extremely rapidly, without any,
really, safeguards, whether we are talking about commercial use or
government use,'' Mr. Cummings said at a hearing on the issue last
month. ``There are real concerns about the risks that this technology
poses to our civil rights and liberties, and our right to privacy.''
Ms. Jackson Lee. First of all, let me say to all of you,
let me thank you for your service to the Nation. I have had the
privilege of serving on this committee for a very, very long
time.
Mr. Wagner, I will get to the underlying basis of this
hearing. But let me be very clear that I have to speak with
great ire and dismay for the behavior of individuals at the
border and the refusal of the Department of Homeland Security
to cooperate with Members of Congress.
I want to indicate that the $4.6 billion that was given
last week and the whining that went on for a period of time to
blame Congress was a misrepresentation to the American people.
Because we understand that reprogramming of dollars can
happen at the drop of a hat. The reason why I say that is, as I
go into my questioning regarding the facial recognition, unless
the answer changed from the time I was here, I understand there
is no statutory legislation or anything that is giving you that
authority. You are going to look for it. Maybe you will answer
that question differently.
But I just quickly want to say that we will not be able to
tolerate--we respect you as servants of the Nation. It is
unfortunate that very destructive policies of this
administration has tainted very fine American servants of the
people. That is what happened. Because when you don't have
toothpaste and a toothbrush and you have a truckload of that
material or nonprofits like the conscious presence that I met
at the border station, one, and also Clint, begging to be of
help, and you are telling the American people there is no one
helping you, I think it is a sad commentary.
So I just want to make sure you are aware of my dismay,
that will not be tolerated, and the mismanagement will not be
tolerated, and the accusations against Members will not be
tolerated.
If Vice President Pence can go in and look after it is
cleaned up, spic and span, then Members who have oversight
responsibility should be able to go in and look.
Mr. Wagner. Understood.
Ms. Jackson Lee. I would appreciate it if you would report
that back to the Secretary.
Mr. Wagner. I will.
Ms. Jackson Lee. Thank you. Let me say to the gentleman
from Transportation Security Administration, I am interested in
you looking into the treatment of Crystal Lynette Sonea and
Sharif Mohamad Hotef--we will give you that information--around
April 14 in the Atlanta airport.
So let me start with Mr. Wagner. This is horrific, the
information regarding the use of these--and my earlier
information was that you know that people of color and women--
so I get it twice--are unfortunately targeted the most.
In the article, it says agents with the FBI and ICE have
turned the State driver's license databases recognition into a
gold mine scanning through hundreds of millions of American
photos without their knowledge or consent.
In addition, it says that the State department motor
vehicle databases into the bedrock of unprecedented
surveillance and infrastructure.
I want to submit into the record, Mr. Chairman, an article
by Amazon that says ``Amazon facial recognition''--not by
Amazon--``Amazon facial recognition mistakenly confused 28
Congresspersons with known criminals''. I will not put the
Congressperson's names into the record, but I think most of us
would like not to be known as known criminals.
My question----
Mr. Payne. No objection.
[The information follows:]
Amazon's Facial Recognition Wrongly Identifies 28 Lawmakers, A.C.L.U.
Says
By Natasha Singer, July 26, 2018, New York Times.
Representative John Lewis of Georgia and Representative Bobby L.
Rush of Illinois are both Democrats, members of the Congressional Black
Caucus and civil rights leaders.
But facial recognition technology made by Amazon, which is being
used by some police departments and other organizations, incorrectly
matched the lawmakers with people who had been charged with a crime,
the American Civil Liberties Union reported on Thursday morning.
The errors emerged as part of a larger test in which the civil
liberties group used Amazon's facial software to compare the photos of
all Federal lawmakers against a database of 25,000 publicly available
mug shots. In the test, the Amazon technology incorrectly matched 28
Members of Congress with people who had been arrested, amounting to a 5
percent error rate among legislators.
The test disproportionally misidentified African-American and
Latino Members of Congress as the people in mug shots.
``This test confirms that facial recognition is flawed, biased and
dangerous,'' said Jacob Snow, a technology and civil liberties lawyer
with the A.C.L.U. of Northern California.
On Thursday afternoon, three of the misidentified legislators--
Senator Edward J. Markey of Massachusetts, Representative Luis V.
Gutierrez of Illinois and Representative Mark DeSaulnier of California,
all Democrats--followed up with a letter to Jeff Bezos, the chief
executive of Amazon, saying there are ``serious questions regarding
whether Amazon should be selling its technology to law enforcement at
this time.''
In the letter, the lawmakers asked for details on how Amazon tested
its facial technology for accuracy and bias. They also requested a list
of all government agencies using Amazon's facial technology as well as
all law enforcement and intelligence agencies Amazon had communicated
with about the system.
Separately, two other congressmen wrongly matched with mug shots--
Mr. Lewis and Representative Jimmy Gomez, a California Democrat--wrote
their own letter to Mr. Bezos requesting an immediate meeting ``to
discuss how to address the defects of this technology.'' The letter was
first obtained by BuzzFeed.
Nina Lindsey, an Amazon Web Services spokeswoman, said in a
statement that the company's customers had used its facial recognition
technology for various beneficial purposes, including preventing human
trafficking and reuniting missing children with their families. She
added that the A.C.L.U. had used the company's face-matching
technology, called Amazon Rekognition, differently during its test than
the company recommended for law enforcement customers.
For one thing, she said, police departments do not typically use
the software to make fully autonomous decisions about people's
identities. ``It is worth noting that in real-world scenarios, Amazon
Rekognition is almost exclusively used to help narrow the field and
allow humans to expeditiously review and consider options using their
judgment,'' Ms. Lindsey said in the statement.
She also noted that the A.C.L.U had used the system's default
setting for matches, called a ``confidence threshold,'' of 80 percent.
That means the group counted any face matches the system proposed that
had a similarity score of 80 percent or more. Amazon itself uses the
same percentage in one facial recognition example on its site
describing matching an employee's face with a work ID badge. But Ms.
Lindsey said Amazon recommended that police departments use a much
higher similarity score--95 percent--to reduce the likelihood of
erroneous matches.
Facial recognition--a technology that can be used to identify
unknown people in photos or videos without their knowledge or
permission--is fast becoming a top target for civil liberties groups
and privacy experts.
Proponents see it as a useful tool that can help identify
criminals. It was recently used to identify the man charged in the
deadly shooting at The Capital Gazette's newsroom in Annapolis, Md.
But civil liberties groups view it as a surveillance system that
can inhibit people's ability to participate in political protests or go
about their lives anonymously. This month, Microsoft said the
technology was too risky for tech companies to deploy without
government oversight and called on Congress to regulate it.
Over the last 2 months, Amazon has come under increasing pressure
for selling its Rekognition technology to law enforcement agencies. The
company has sold the service as a way for police departments to easily
identify suspects in photos or videos.
Amazon's site describes how its system can perform ``real-time face
recognition across tens of millions of faces'' and detect ``up to 100
faces in challenging crowded photos.'' (The New York Times recently
used the Amazon technology to help identify guests at the royal wedding
of Prince Harry and Meghan Markle.)
In May, two dozen civil liberties groups, led by the A.C.L.U.,
wrote a letter to Mr. Bezos, demanding that his company stop selling
the facial technology to law enforcement. The groups warned that the
software could be used to trail protesters, undocumented immigrants or
other members of the public--not just criminal suspects.
Similar demands of Mr. Bezos from Amazon employees, Amazon
investors, and several hundred academics soon followed.
Mr. Snow of the A.C.L.U. said his group's test of Amazon's software
should push Congress to put a moratorium on law enforcement's use of
facial recognition technology.
But in a blog post last month, Matt Wood, general manager of
artificial intelligence at Amazon Web Services, said that there had
been no reports of law enforcement abuse of Amazon's facial technology.
He added that Amazon believed it was ``the wrong approach to impose a
ban on promising new technologies because they might be used by bad
actors for nefarious purposes in the future.''
In a letter to Amazon, the Congressional Black Caucus noted the
potential for racial bias with the technology--an issue raised by a
recent M.I.T. study that found some commercial facial recognition
systems correctly identified a higher proportion of white men than
darker-skinned women. In their letter, the caucus members urged Mr.
Bezos to hire ``more lawyers, engineers and data scientists of color to
assist in properly calibrating this technology to account for racial
bias that can lead to inaccuracies with potentially devastating
outcomes.''
In the civil liberties group's test, the Amazon software
misidentified several members of the Congressional Black Caucus,
including Mr. Lewis and Mr. Rush, as other people who had been
arrested.
``We think these test results really raise the concern that facial
recognition has a race problem,'' said Mr. Snow, the A.C.L.U. lawyer.
Ms. Jackson Lee. To both of you, and a little extra time
for them to answer, the two gentlemen from TSA and from CBP,
how are you doing this, with the protections of due process and
notice, without the notice of the American people that the
process even exists? What framework is there to have the
firewalls that you are not turning Congress people or children
into convicted criminals?
Mr. Wagner. We are not seeing those same error rates that
are--that can be attributed to specific demographics in how we
are doing this.
How we are doing this cannot be compared to previous
studies on this. There are different control factors in place.
You know, there are different--we are taking a person that is
standing in front of a camera where we can take a clear
picture, and we are comparing it against a clear set of
baseline photos from their passports or their visas where they
were also standing still in front of a camera to capture a
clear picture. That is why we have such accurate rates.
Previous studies didn't quite take the same control factors
into place. This is not us taking an image of a person and
randomly running it against a gallery set of indistinguishable,
say, quality photographs and lowering down the accuracy rate as
to what constitutes a match to make it match someone that it is
not.
I mean, you can do the same thing with fingerprints. If you
only take two----
Ms. Jackson Lee. How do you secure that--how do you secure
that data?
Mr. Wagner. When the photo is taken at the airport, it is
encrypted and transmitted to the CBP into our cloud space. It
is then templatized, which my understanding of that is it is
turned into a mathematical formula. There is a unique
identifier associated with that. There is no biographical data
or PII associated with that. It is matched up against our
gallery of templatized photos. When there is a match, a message
goes back to the camera with just yes or no and that unique
identifier.
Ms. Jackson Lee. Let me move quickly to Mr. Gould and TSA.
Let me thank TSA for their front-line service of protecting
America.
Mr. Payne. Thank you, Ms. Lee.
Ms. Jackson Lee. Thank you, Mr. Chairman, for your
indulgence.
The same question as to how you are utilizing and how you
are protecting the data and avoiding this intrusion into the
privacy of the American public without them knowing it.
Mr. Gould. Yes, ma'am. So I would--we are using CBP's TVS
system. So the answer that Mr. Wagner provided applies to TSA
as well.
With respect to the accuracy and the matching, the one
thing that I would like to add is, the technology is evolving
so quickly and it is improving so quickly, we will continue to
assess at every step, for any additional pilots, from when we
consider employing this in a wider scale, we will assess the
best way to get quality image capture and be sure to employ the
highest-quality algorithms to ensure the highest match rate.
Mr. Payne. OK. Thank you.
Ms. Jackson Lee. Thank you, Mr. Chairman.
I yield back. Thank you very much.
Mr. Payne. The Chair recognizes Mr. Green.
Mr. Green of Texas. Thank you, Mr. Chairman.
Thank the Ranking Member. Thank the witnesses for
appearing.
My questions have to do with the surveillance. My first
question is, are all people who are traversing areas within an
airport under some degree of suspicion?
Who would like to answer, please?
Mr. Gould. Well, I would say that when a person is
traversing an airport, they are not necessarily under
suspicion. Airports, you know, utilize security cameras, closed
circuit television, for security reasons.
With respect to TSA, though, the only reason that we use
cameras and capture images is solely for the purpose of
identification.
Mr. Wagner. If I could just add that----
Mr. Green of Texas. Please.
Mr. Wagner [continuing]. What we are doing is absolutely
not a surveillance program. The picture of an individual is
taken with their complete knowledge, because they are standing
in front of a camera at a time and place where they have to
present a physical ID in order to establish their identity to
move forward. We are just replacing the evaluation and the
scrutiny of the physical ID with a computer algorithm.
Mr. Green of Texas. Should I assume that persons who enter
the airport and who are not within the secured area will not be
a subject of this technology?
Mr. Gould. Not by TSA, sir. It solely occurs at either the
bag drop or the checkpoint.
Mr. Wagner. Or a time and place where you have to present
an identification to establish your identity to go through
whatever process that is.
Mr. Green of Texas. In Houston, the bag drop occurs outside
of the building, before you enter the building. You drive up in
your car, you have friends, neighbors with you perhaps, and you
go over to an agent, and that person then receives your bag,
gives you a ticket.
So would it occur in this area, please?
Mr. Gould. Sir, right now, the only place that the
biometric identification that the bag drop is occurring is in
Terminal F in Atlanta. I went down there. I observed the way
the technology----
Mr. Green of Texas. If I may, because time is of the
essence. But we are talking about expanding, are we not?
Mr. Gould. Yes, sir.
Mr. Green of Texas. OK. Here is my concern. Let me go to
the point, and I will be as pithy as I can.
But one can only imagine what Mr. J. Edgar Hoover would
have done with this technology. It was Mr. Hoover who
surveilled Dr. King. They went so far as to send a letter to
Dr. King encouraging him to take his life. One can only
imagine.
Now, I am not placing you under the eye of suspicion, but
it is my job to make sure that this kind of technology is not
abused. I take my job seriously because I am protecting you by
doing my job.
So my concerns are, do you alert people in some way to--so
as to advise them that they are being surveilled?
Mr. Gould. Sir, I wouldn't characterize it as surveillance.
The way the alert happens, to use your term, is when you
approach the bag drop, the agent will say, ``Would you like to
use biometric identification to match you to your bag?'' or
something along those lines.
Mr. Green of Texas. Permit me to ask this. If you thought--
if you believed that this was a form of surveillance, would you
alert people? Would you alert the public, if you thought this
was some form of surveillance?
Mr. Gould. So we don't do surveillance, but we are----
Mr. Green of Texas. Excuse me. If you thought--would you
recommend--if we were of the opinion that this is surveillance,
what do you think we should do? Should we indicate that person
should be noticed that they are being surveilled?
Mr. Gould. Sir, we provide notice before the image is
captured. It is purely with the consent of the traveler.
Mr. Green of Texas. What about the consent of the person
who happens to be with the traveler who is just a friend?
Mr. Gould. We solely capture the picture of the traveler
who has consented. The camera is only about 2 feet away. You
step right in front of it, and it solely captures that image.
Mr. Green of Texas. All right. Thank you. But we are
considering expansion.
My concern is suspicionless surveillance, suspicionless
surveillance, surveilling persons who are not under suspicion
perhaps by accident.
Final question is this because time is running out.
Will there be any means by which persons who engaged in
litigation can acquire access to this intelligence that you
have preserved for some length of time, meaning the
photographs?
Will there be any means by which persons who are engaged in
litigation can acquire it?
Mr. Gould. Sir, the photographs we match against are in the
CBP TVS system. They are passport photographs. The images that
are captured are not retained in the camera in any respect. We
solely get back a match/no match return, if that answers your
question.
Mr. Green of Texas. It really does not, because what I am
trying to get to is this: If persons are engaged in some form
of litigation--and one can only imagine what that might be--
will they be able to acquire a photo so as to show that a
person was at a given location on a given occasion?
Mr. Gould. I understand, sir. That photo is not retained at
all by TSA, so they will not be maybe to retain it.
Mr. Green of Texas. It is retained----
Mr. Gould. It is encrypted. It is transmitted to CVP, and a
match rate is returned.
Mr. Green of Texas. OK.
Mr. Wagner. If it is a U.S. citizen, the photo is deleted
after 12 hours. If it is a foreign national, at the baggage
drop, that photo would also be deleted. What we would keep on a
foreign national, though, is their boarding on the plane and
their final departure to serve as the biometric exit of their
departure.
Mr. Green of Texas. Thank you. I greatly appreciate this. I
assure you that I want us to secure our airports, our ports of
entry, but I am also concerned about suspicionless
surveillance.
Thank you.
Mr. Payne. Thank you.
The gentleman, Mr. Guest, you are recognized for 5 minutes.
Mr. Guest. Thank you, Mr. Chairman.
Mr. Wagner and other guests, thank you for being here
today. I know that at least 3 of our witnesses, your
departments fall under the Department of Homeland Security.
Your website reads as follows: The Department of Homeland
Security has a vital mission: To secure the Nation for the many
threats we face. This requires the dedication of more than
240,000 employees in jobs that range from aviation and border
security to emergency response, from cybersecurity analysts to
chemical facility inspectors. Our goal is clear: Keeping
America safe.
In addition to the agencies that are represented here
today, Homeland Security includes the Cybersecurity and
Infrastructure Security Agency, the United States Citizen and
Immigration Services, the United States Coast Guard, the United
States Immigration and Customs Enforcement. It includes FEMA as
well as the Customs and Border Protection, Secret Service, and
the TSA.
I believe that if these agencies that I just spoke of, if
these agencies were abolished, that our country would be
substantially less safe.
My question--I will begin with you, Mr. Wagner--is can you
please tell me what impact it would have on the people of
America if Homeland Security and these agencies for which you
serve, if these agencies were abolished by Congress?
Mr. Wagner. Well, there would be no one to process people
coming and going across the border, either U.S. citizens or
visitors. There would be no one to process commercial cargo, to
look for harmful goods or products coming in. There would be no
one to collect the taxes that are due on those duties. CBP
collects over $40 billion a year into the U.S. Treasury through
duties, taxes, and fees. There would be no one to do that.
Mr. Guest. Would you agree with me that the different
enforcement capacities that the Department of Homeland Security
polices, that it runs a gamut of different things? We just
talked about everything from the Secret Service, which provides
protection for our dignitaries; TSA, which is responsible for
air travel; Coast Guard; border enforcement--that those are
very important functions of our Government to make sure that
these agencies are funded? Would you agree with that, Mr.
Wagner?
Mr. Wagner. Yes. The origins of our agency go back to 1789
and the very beginning of the country.
Mr. Guest. Mr. Gould, would you care to expound on that at
all?
Mr. Gould. I agree with what Mr. Wagner said, and, you
know, if TSA were not there, the security of transportation
systems, not solely air travel, would be in some degree of
jeopardy.
Mr. Di Pietro. Congressman, as you indicated, you know, we
protect the President and the Vice President, others. We also
have criminal investigations. So that is critical work that we
are doing.
Mr. Guest. Would each of you agree that it would be
irresponsible to talk about abolishing these agencies that
perform such very important tasks on behalf of the American
people?
Mr. Wagner. Yes.
Mr. Gould. Yes, sir.
Mr. Di Pietro. I would agree with that.
Mr. Guest. No further questions, Mr. Chairman. I yield
back.
Mr. Payne. Thank you. I, you know, just--I know I was late
to the hearing today, but I don't really--and maybe it happened
before I got here, but I don't really ever mention--hearing
anyone mention that these institutions should be abolished, so
just for the record.
We have the gentleman from Kansas City, Mr. Cleaver.
Mr. Cleaver. Thank you, Mr. Chairman.
I am deviating a little. Do you know James Wilson? Do any
of you know who he is? Probably one of the most important
figures that we don't know much about. He signed the
Declaration of Independence and eventually became a member, 1
of the first 6 members of the Supreme Court. He said that the
Congress shall form the grand inquisition of the Executive
branch. I think that my children's children, and even their
children, will study this era and say: That is when it got
started.
I am concerned, you know, I was in the executive branch
municipality, mayor of Kansas City, and so I know you guys are
busy, especially right now. A group of my colleagues and I
signed a letter and sent it to you, Mr. Wagner and Mr. Gould,
almost 30 days ago. We haven't gotten an answer. So I didn't
know if this was a part of the plan to ignore Congress or if
you are just consumed. I am not stupid, so I know you don't
have--nobody should expect you to write a personal letter to
everybody who writes you a letter, even Members of Congress,
but if you don't have enough staff, we need to know. Because
until it completely collapses, we are still supposed to provide
oversight. I am not trying to be hostile. I am not sure I can
do a good job of being hostile, but I can certainly do a good
job of being frustrated. So I appreciate your work and what you
do, but I just--I have to say that it is frustrating, just
listening, just seeing what is going on, refusal after refusal
to allow Congress to do its oversight. I hope that if I am
around at a time when my voice is important, to say, I am not
going to support nonresponsiveness to Congress, that I get the
opportunity to say it, even if my daddy is in the White House.
Now, having said that, some of the questions that my
colleagues and I asked because we thought they were important,
I will ask a couple of them. Time is running out, but is there
any statutory authority that would allow the whole process of
facial recognition, or is that just an internal move? Anybody?
Mr. Wagner. There are several pieces of statutory authority
that authorize us to do and run this program. There are several
pieces of legislation from Congress, requiring a biometric-
based, entry-exit system for certain foreign nationals. There
are other statutes which authorize us to determine identity and
citizenship, including U.S. citizens. There has to be a way for
us to make that determination that a person is a U.S. citizen,
and there are statutes to authorize us to consider evidence
presented by that person to make that determination, and then
if it is not to the examining officer's satisfaction, the
regulations stipulate that person would be considered and
inspected as an alien.
Mr. Cleaver. OK. Thank you.
Mr. Gould.
Mr. Gould. Sir, from a TSA perspective, the Aviation and
Transportation Security Act requires that we screen all
passengers and crew boarding aircraft. Fundamental in that
screening process is that we positively identify them. The Act
mentions exploring the use of biometrics for that purpose. So
that is the authority that we are operating under.
Mr. Cleaver. OK. I mean, it wasn't a trick question. I just
wanted to know.
Mr. Gould. No, I understand, sir.
Mr. Cleaver. Yes. Last week, I participated in a
demonstration in front of the Treasury Department, along with a
number of other individuals, the refusal to put a
Congressionally-approved likeness of an African American woman
on the dollar. That is another whole issue, but I was in front
of the demonstration. Should I and the other folk who got off
that bus to demonstrate expect that we were somehow surveilled
and put in the category of subjects of interest? I mean, since
that is what apparently takes place on the grounds of the White
House. I don't want to suggest I am as important as, you know,
the President or Patrick Mahomes or somebody, but, you know,
should I expect that?
Mr. Di Pietro. Congressman, we do have a CCTV video
surveillance system in and around the White House. There is a
PIA that is published through the Department of Homeland
Security alerting people to that. In addition, the cameras that
we have, many of them are overt, all down Pennsylvania Avenue
and on the buildings adjacent to the White House there.
Mr. Cleaver. What about other Federal departments?
Mr. Di Pietro. I can't speak to what other Federal
Departments are doing, Congressman.
Mr. Cleaver. OK. All right. Thank you.
Mr. Gould, nice suit.
Mr. Gould. Thank you very much, sir. I like yours, too.
Mr. Cleaver. I yield back, Mr. Chair.
Mr. Payne. Thank you, sir. Now we recognize the gentlelady
from Florida, Mrs. Demings.
Mrs. Demings. Thank you so much, Mr. Chairman, and thank
you to our witnesses today. Let me just for the record say that
I respect the jobs that you have to do. I understand how tough
they are. I think that all of our jobs have gotten tougher in
recent years. I am not sure why my colleague felt the need to
talk about abolishing your agencies, because I know no one on
this committee, on either side of the aisle, has ever proposed
such an idea. We are the Committee on Homeland Security, and we
are here to make sure that you have the tools and resources to
effectively do your jobs, but I know that gets a little tougher
when sometimes you receive unjust and improper orders and do
not have the resources to effectively do your job.
Earlier I heard one of my colleagues talk about the reason
for biometric technologies involved speed and efficiency. Well,
I was assigned to the Orlando International Airport as a police
commander on the worst day in aviation history, on 9/11. I know
that the No. 1 responsibility for you is the safety of the
traveling public, and if you can ensure that, or increase those
odds and do it in an efficient and faster way, then that is
just icing on the cake.
But what sets us apart as we work to keep our Nation safe,
what sets us apart in this country is that we can enforce the
laws and write the laws, but also protect an individual's civil
rights. That is what sets us apart. I will not--violating civil
rights or the perception of violating civil rights is an issue
that we cannot ignore and we have to deal with. Look, when we
are able to deploy new technology, that is a great and
wonderful thing. I remember how exciting that was, but it is
our job, on the committee, and your jobs, as the head of your
agencies, to make sure that we can do it all. I believe in this
Nation we can.
I know we have talked about every different thing that we
possibly could. We do thank you for your endurance. I just want
to go back for just a minute to testing for accuracy and any
biases. Could you tell me who sets the minimum standards for
this particular program, like, who decides what testing is done
for accuracy or bias, is conducted before deploying the
technology? How do you get that baseline and say that this
technology, we have done the testing, we have spoken to the
stakeholders, we are ready for prime time now? Understanding,
as I believe you said earlier, that we are always fine-tuning
and going back and checking up, but who sets the original kind-
of standards before deployment? What is acceptable and
unacceptable? Mr. Wagner, we will start with you.
Mr. Wagner. Sure. So we would do that internally. We would
determine what constitutes a match versus a nonmatch to a
photo. We would evaluate this with our DHS Science and
Technology Department. We would do it in consultation with
NIST. We do it in consultation with experts from the industry
and the vendors of this equipment. We have partnered with NIST,
and starting this summer into the fall, we will be deeply
analyzing the results of our data to make sure that we are not
seeing those error rates that are attributable to a certain
demographic. We are not seeing it from our internal review of
it, but we want to make sure, so we are bringing the experts in
to make sure----
Mrs. Demings. Right. So you are saying it is a perception
that there is an increased error rate among people of color, or
have we seen some data, although not significant, to show that?
Mr. Wagner. I think the studies that have shown there were
these biases in it had different control factors than how we
are using this program. No one has really studied the way that
we are implementing this using those same control factors on
how we are doing it, and I would expect them to get the similar
results as to we are seeing.
Mrs. Demings. Mr. Gould, can you----
Mr. Gould. Ma'am, from a TSA perspective, we work very
closely with the DHS Science and Technology Director as well.
They inform our test plans and how we collect data on the
biometric pilots and how well they are working, and then they
analyze that data on our behalf. So we really do rely on them
for their semi-independent and very, very accurate assessments
of our capability.
Then, like CBP, we rely on our friends at NIST as well to,
you know, really set the standards and say how well the
algorithms are actually working.
Mrs. Demings. So when you decide--Mr. Chairman, if I could
just--when you decide that this--we are ready for deployment,
this technology, based on the testing we have done, is ready
for prime time, who makes that decision? Is it a collective
effort between the different people that you work with, or do
you decide that individually based on the feedback that you
receive?
Mr. Wagner. We would decide that for our agency, because it
is our responsibility. The officer's determination, you match
your passport, and if I use a tool or an algorithm to help me
make that decision, at the end of the day, it is still my
judgment to do that. So we would evaluate this to say, is this
helpful to the officer making that determination, that this
document corresponds to that person.
Mrs. Demings. OK.
Mr. Gould. One thing I would add to your original point,
for us, the main reason to do this is increased--better
identity verification, right, and the secure enhancements that
are associated with that. Getting people through the checkpoint
more quickly, like you said, is kind of icing on the cake. But
better security through using this technology is really, really
key to us. If the algorithms and the match rates are not
acceptable, if we are not enhancing security, then we will not
deploy it, but that decision would be made internal at the TSA.
Mrs. Demings. Thank you, Mr. Chair.
I yield back.
Mr. Payne. I thank the gentlelady, and I just--probably due
to the time, I will dispense with my questions, but I would
just like to say that obviously based on the questioning from
the Members of Congress, you can get a feeling on where we are
concerned about issues around privacy, around equality, and
making sure that the American people and the traveling public
is safe. So we need to continue to evolve, and we know that
Homeland Security has been an evolving, living, breathing
entity that continues to have to see and recognize issues, try
to curtail them, and rectify matters that are important to the
American people.
So I would just like to say, thank you for your service in
TSA, CBP. Your jobs, all of you actually, Secret Service, are
doing a yeoman's job for this Nation, and we appreciate your
service and your time here today, so thank you.
With that, the hearing is adjourned.
[Whereupon, at 12:18 p.m., the committee was adjourned.]