PDF(PDF provides a complete and accurate display of this text.)Tip?
Calendar No. 197
116th Congress } { Report
SENATE
1st Session } { 116-93
_______________________________________________________________________
DEEPFAKE REPORT ACT OF 2019
__________
R E P O R T
OF THE
COMMITTEE ON HOMELAND SECURITY AND
GOVERNMENTAL AFFAIRS
UNITED STATES SENATE
TO ACCOMPANY
S. 2065
TO REQUIRE THE SECRETARY OF HOMELAND SECURITY TO PUBLISH AN ANNUAL
REPORT ON THE USE OF DEEPFAKE TECHNOLOGY, AND FOR OTHER PURPOSES
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
September 10, 2019.--Ordered to be printed
________
U.S. GOVERNMENT PUBLISHING OFFICE
89-010 WASHINGTON: 2019
_____________________________________________________________________________
COMMITTEE ON HOMELAND SECURITY AND GOVERNMENTAL AFFAIRS
RON JOHNSON, Wisconsin, Chairman
ROB PORTMAN, Ohio GARY C. PETERS, Michigan
RAND PAUL, Kentucky THOMAS R. CARPER, Delaware
JAMES LANKFORD, Oklahoma MAGGIE HASSAN, New Hampshire
MITT ROMNEY, Utah KAMALA D. HARRIS, California
RICK SCOTT, Florida KYRSTEN SINEMA, Arizona
MICHAEL B. ENZI, Wyoming JACKY ROSEN, Nevada
JOSH HAWLEY, Missouri
Gabrielle D'Adamo Singer, Staff Director
Joseph C. Folio III, Chief Counsel
Michelle D. Woods, Co-Director and Chief Policy Advisor for Homeland
Security
David M. Weinberg, Minority Staff Director
Zachary I. Schram, Minority Chief Counsel
Michelle M. Benecke, Minority Senior Counsel
Jeffrey D. Rothblum, Minority Fellow
Laura W. Kilbride, Chief Clerk
Calendar No. 197
116th Congress } { Report
SENATE
1st Session } { 116-93
======================================================================
DEEPFAKE REPORT ACT OF 2019
_______
September 10, 2019.--Ordered to be printed
_______
Mr. Johnson, from the Committee on Homeland Security and Governmental
Affairs, submitted the following
R E P O R T
[To accompany S. 2065]
[Including cost estimate of the Congressional Budget Office]
The Committee on Homeland Security and Governmental
Affairs, to which was referred the bill (S. 2065) to require
the Secretary of Homeland Security to publish an annual report
on the use of deepfake technology, and for other purposes,
having considered the same, reports favorably thereon with an
amendment (in the nature of a substitute) and recommends that
the bill, as amended, do pass.
CONTENTS
Page
I. Purpose and Summary..............................................1
II. Background and Need for the Legislation..........................1
III. Legislative History..............................................4
IV. Section-by-Section Analysis......................................4
V. Evaluation of Regulatory Impact..................................5
VI. Congressional Budget Office Cost Estimate........................5
VII. Changes in Existing Law Made by the Bill, as Reported............6
I. PURPOSE AND SUMMARY
S. 2065, the Deepfake Report Act of 2019, requires the
Secretary of Homeland Security to publish an annual report on
the extent digital content forgery technologies, also known as
deepfake technologies, are being used to weaken national
security, undermine our nation's elections, and manipulate
media.
II. BACKGROUND AND NEED FOR LEGISLATION
Advances in machine learning algorithms and artificial
intelligence capabilities have accelerated the proliferation of
digital content forgery technologies, commonly referred to as
``deepfake technologies.''\1\ Deepfake technologies are used to
manipulate audio, video or other media content and have the
potential to be used to undermine national security, erode
public trust in our democracy and other nefarious reasons.\2\
As the software underpinning these technologies becomes easier
to acquire and use, policy makers and national security experts
are concerned that the continued dissemination of deepfake
content across trusted media platforms could increasingly be
used to dupe audiences and amplify false narratives about
American cultural norms and interests domestically and
abroad.\3\
---------------------------------------------------------------------------
\1\Kristina Libby, This Bill Hader Deepfake Video is Amazing. It's
Also Terrifying for Our Future, Popular Mechanics (Aug. 13, 2019),
https://www.popularmechanics.com/technology/security/a28691128/
deepfake-technology/.
\2\Id.
\3\Cuihua Shen, Mona Kasra, Wenjing Pan, Grace A. Bassett, Yining
Malloch, and James F. O'Brien, Fake Images: The Effects of Source,
Intermediary, and Digital Media Literacy on Contextual Assessment of
Image Credibility Online, New Media & Society, Vol. 21, 2:pg. 238-463,
SAGE Publications (2018), available at http://graphics.berkeley.edu/
papers/Shen-FIT-2018-09/Shen-FIT-2018-09.pdf [hereinafter Fake Images],
The National Security Challenges of Artificial Intelligence,
Manipulated Media, and `Deepfakes': Hearing Before the U. S. House of
Representatives, Permanent Select Committee on Intelligence, 116th
Cong., (2019) [hereinafter HPSCI Hearing] (statement of Clint Watts).
---------------------------------------------------------------------------
Doctored videos of high-profile politicians and Facebook
founder Mark Zuckerberg are prime examples of the disturbing
impacts of the use of digital content forgeries.\4\ The threat
of weaponizing information, even without this powerful
technology, becomes clear when considering the Russian
influence campaigns carried out in recent U.S. elections.\5\
The Intelligence Community concluded that Russia's Internet
Research Agency placed false social media advertisements and
manipulated content across various highly trafficked and
trusted media platforms leading up to the 2016 and 2018 mid-
term elections.\6\
---------------------------------------------------------------------------
\4\Allyson Chiu, Facebook Wouldn't Delete An Altered Video of Nancy
Pelsoi. What about one of Mark Zukerberg?, The Washington Post (June
12, 2019), https://www.washingtonpost.com/nation/2019/06/12/mark-
zuckerberg-deepfake-facebook-instagram-nancy-pelosi/.
\5\Susannah George, ``Deepfakes'' Called New Election Threat, With
No Easy Fix, Associated Press (June 13, 2019), https://www.apnews.com/
4b8ec588bf5047a981bb6f7ac4acb5a7.
\6\HPSCI Hearing, supra note 3; Madeline Purdue, Deepfake 2020: New
Artificial Intelligence is Battling Altered Videos Before Elections,
USA Today (Aug. 14, 2019), https://www.usatoday.com/story/tech/news/
2019/08/14/election-2020-company-campaigns-against-political-deepfake-
videos/2001940001/.
---------------------------------------------------------------------------
In advance of the 2016 and 2018 mid-term elections,
Facebook estimated that from January 2015 to August 2017,
Russian-backed bots spread fabricated media content to about
half of the 250 million eligible voters.\7\ Moreover, Twitter
discovered approximately ``tens of thousands automated
accounts'' linked directly to Russia.\8\ In an effort to limit
the flow of disinformation on its platform, Twitter reportedly
suspended more than 70 million accounts linked to Russian bots
in May and June of 2018 and continued to do so throughout
2018.\9\
---------------------------------------------------------------------------
\7\Oren Etzioni, How Will We Prevent AI-Based Forgery, Harvard
Business Review (March 1, 2019), https://hbr.org/2019/03/how-will-we-
prevent-ai-based-forgery.
\8\Id.
\9\Craig Timberg and Elizabeth Dwoskin, Twitter is Sweeping Out
Fake Accounts Like Never Before, Putting User Growth at Risk, The
Washing Post (July, 16, 2018), https://www.washingtonpost.com/
technology/2018/07/06/twitter-is-sweeping-out-fake-accounts-like-never-
before-putting-user-growth-risk/.
---------------------------------------------------------------------------
As cyber-enabled warfare increasingly becomes the norm,
national security experts warn that if the Federal Government
does not take swift action to address persistent purveyors of
information warfare, deepfake technologies will only continue
to become more sophisticated and widely used in disinformation
campaigns launched by our nation's foreign adversaries, most
notably China and Russia.\10\ In June 2019, the U.S. House of
Representatives Permanent Select Committee on Intelligence held
a hearing entitled, ``The National Security Challenges of
Artificial Intelligence, Manipulated Media, and
`Deepfakes'''.\11\ During this hearing, witness Clint Watts,
Distinguished Research Fellow, Foreign Policy Institute, stated
that China and Russia will continue to use deepfake
technologies to ``discredit domestic dissidents and foreign
detractors, incite fear and promote conflict inside Western-
style democracies, and distort the reality of American
audiences and audiences of American allies.''\12\ Mr. Watts
further explained the dangers of the continued proliferation of
deepfake technologies:
---------------------------------------------------------------------------
\10\HPSCI Hearing, supra note 3 (statement of Clint Watts); Fake
Images, supra note 3.
\11\HPSCI Hearing, supra note 3.
\12\Id. (statement of Clint Watts).
Over the long term, deliberate development of false
synthetic media will target U.S. officials,
institutions and democratic processes with an enduring
goal of subverting democracy and demoralizing the
American constituency. In the near and short term,
circulation of ``Deepfakes'' may incite physical
mobilizations under false pretenses, initiate public
safety crises and spark the outbreak of violence. The
recent spate of false conspiracies proliferating via
WhatsApp in India offer a relevant example of how bogus
messages and media can fuel violence. The spread of
``Deepfake'' capabilities will likely only increase the
frequency and intensity of such violent outbreaks.\13\
---------------------------------------------------------------------------
\13\Id.
Federal entities, such as Defense Advanced Research
Projects Agency, and academic institutions are conducting
research on how to identify doctored media and counter deepfake
technologies.\14\ However, national security experts have
identified a series of interim actions Congress and the private
sector can take to counter the proliferation of deepfake media
content and its adverse effects. Proposed actions to counter
deepfake technologies and content include: implementing
legislation prohibiting public figures from disseminating
deepfake content, public-private partnerships to develop
standards for content accountability, encouraging social media
companies to develop digital signatures, and launching public
awareness campaigns about deepfake content.\15\
---------------------------------------------------------------------------
\14\Kristina Libby, This Bill Hader Deepfake Video is Amazing. It's
Also Terrifying for Our Future, Popular Mechanics, (Aug. 13, 2019),
https://www.popularmechanics.com/technology/security/a28691128/
deepfake-technology/.
\15\Id.; HPSCI Hearing, supra note 3 (statement of Clint Watts);
Fake Images, supra note 3.
---------------------------------------------------------------------------
The Federal Government should proactively identity the
tools and techniques used by our adversaries to develop
deepfake technologies and content, and develop countermeasures
and tools to identify and counter deepfake content.
S. 2065 requires the Department of Homeland Security, in
coordination with other Federal agencies, to develop a report
on the use of digital content forgery technologies, and
recommend actions and identify countermeasures to mitigate the
impacts of these technologies on national security. The report
is to include, among other things, an assessment of the
technologies used to create or propagate digital content
forgeries, a description of the types of digital content
forgeries, how such technologies are being or could be used to
undermine national security, and a description of the
technological countermeasures that are, or could be, used to
address concerns with digital content forgeries. In addition,
the Secretary may consult with other Federal agencies and
conduct public hearings to gather relevant information to
assist with the production of the report.
The first report is due one year after the bill's
enactment, and annually thereafter for five years. The report
is to be provided in unclassified form, but may include a
classified annex. Existing information and public disclosure
exemptions under the Freedom of Information Act apply to the
reports required by this bill. The bill also includes an
exemption from the Paperwork Reduction Act.
III. LEGISLATIVE HISTORY
S. 2065, the Deepfake Report Act of 2019, was introduced on
July 09, 2019, by Senator Rob Portman. Senators Martin
Heinrich, Brian Schatz, Cory Gardner, Mike Rounds, Joni Ernst,
and Gary Peters cosponsored the bill. Senator Margaret Wood
Hassan later joined as a cosponsor. The bill was referred to
the Committee on Homeland Security and Governmental Affairs.
The Committee considered S. 2065 at a business meeting on
July 24, 2019. During the business meeting, a substitute
amendment by Senator Portman was offered and adopted. The
substitute amendment added language to include a sunset
provision, limit the scope of the report to ways that deepfakes
are used to commit fraud, cause harm, and violate federal civil
rights, and clarify applicability of Freedom of Information Act
and Paperwork Reduction Act requirements. Both the amendment
and the legislation as modified were passed by voice vote en
bloc with Senators Johnson, Portman, Paul, Lankford, Romney,
Scott, Enzi, Hawley, Peters, Carper, Hassan, Sinema, and Rosen
present.
IV. SECTION-BY-SECTION ANALYSIS OF THE BILL, AS REPORTED
Section 1. Short title
This section establishes the short title of the bill as the
``Deepfake Report Act of 2019.''
Section 2. Definitions
This section defines ``digital content forgery'' as the use
of emerging technologies, including artificial intelligence and
machine learning techniques, to fabricate or manipulate audio,
visual, or text content with the intent to mislead. This
section defines ``Secretary'' as the Secretary of Homeland
Security.
Section 3. Reports on digital content forgery technology
Subsection (a) requires the Secretary, acting through the
Under Secretary for Science and Technology to produce a report
on the state of digital content forgery technology. The report
is to be provided one year after the enactment of the Act, and
annually thereafter for five years.
Subsection (b) specifies the required contents of the
report. The report is to include: an assessment of the
technologies used to create or propagate digital content
forgeries; a description of the types of digital content
forgeries, including those used to commit fraud, to cause harm,
or to violate civil rights; and an assessment of how foreign
governments, and their proxies and networks, use or could use
digital content forgeries to undermine national security; an
assessment of deep learning technologies used to generate
deepfakes to commit fraud, cause harm, or violate civil rights;
an analysis of the effectiveness of methods used to identify
deep fakes; and, recommendations for employing methods to
identify deepfakes as well as effective public warning methods
for specific content. The report should also include, among
other things as deemed appropriate by the Secretary, a
description of the technological countermeasures that are, or
could be, used to address concerns with digital content
forgeries.
Subsection (c) allows the Secretary to consult with any
other Federal agency or other interested parties and conduct
public hearings to gather information and advice relevant to
complete the reports required by this bill.
Subsection (d) stipulates that the reports should be
unclassified, but may contain a classified annex.
Subsection (e) states that information used to produce the
reports required by this bill that are exempt from the
requirements under section 552 of title 5, United States Code
(commonly known as the ``Freedom of Information Act'') are to
remain exempt.
Subsection (f) states that the reports required by this
bill are exempt from Subchapter I of chapter 35 of title 44,
United States Code (commonly known as the ``Paperwork Reduction
Act'').
V. EVALUATION OF REGULATORY IMPACT
Pursuant to the requirements of paragraph 11(b) of rule
XXVI of the Standing Rules of the Senate, the Committee has
considered the regulatory impact of this bill and determined
that the bill will have no regulatory impact within the meaning
of the rules. The Committee agrees with the Congressional
Budget Office's statement that the bill contains no
intergovernmental or private-sector mandates as defined in the
Unfunded Mandates Reform Act (UMRA) and would impose no costs
on state, local, or tribal governments.
VI. CONGRESSIONAL BUDGET OFFICE COST ESTIMATE
U.S. Congress,
Congressional Budget Office,
Washington, DC, August 16, 2019.
Hon. Ron Johnson,
Chairman, Committee on Homeland Security and Governmental Affairs, U.S.
Senate Washington, DC.
Dear Mr. Chairman: The Congressional Budget Office has
prepared the enclosed cost estimate for S. 2065, the Deepfake
Report Act of 2019.
If you wish further details on this estimate, we will be
pleased to provide them. The CBO staff contact is Aldo
Prosperi.
Sincerely,
Phillip L. Swagel,
Director.
Enclosure.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
S. 2065 would require the Department of Homeland Security
(DHS) to submit five annual reports to the Congress on digital
content forgeries, also known as ``deepfakes.'' Such forgeries
manipulate digital content, such as videos, with the intent to
mislead the viewer. The bill would require DHS to assess the
use of digital content forgeries by foreign entities and
evaluate available methods of detecting and mitigating such
threats.
On the basis of information from DHS and considering
information about similar reporting requirements, CBO estimates
that enacting S. 2065 would cost less than $500,000 over the
2019-2024 period. Such spending would be subject to
availability of appropriated funds.
The CBO staff contact for this estimate is Aldo Prosperi.
The estimate was reviewed by Leo Lex, Deputy Assistant Director
for Budget Analysis.
VII. CHANGES IN EXISTING LAW MADE BY THE BILL, AS REPORTED
Because S. 1151 would not repeal or amend any provision of
current law, it would make no changes in existing law within
the meaning of clauses (a) and (b) of paragraph XXVI of the
Standing Rules of the Senate.
