Through A Straw Darkly: Reflections on the NYU Conference “When Seeing Isn’t Believing: Deepfakes and the Law”
By: Matthew F. Ferraro1Matthew F. Ferraro, an attorney and former U.S. intelligence officer, is a term member of the Council on Foreign Relations and a visiting fellow at the National Security Institute at George Mason Law School. Twitter: @MatthewFFerraro
April 4, 2020
During New York University’s March 2, 2020 conference on deepfakes, the ACLU’s Ben Wizner made a thought-provoking observation. He said that when thinking about disinformation issues, just “[f]ocusing on deepfakes is like looking through a straw.” It is a genuine societal problem, he said, but so are bogus stories propagated online and other forms of “cheap” deception. He cited fake fliers about immigrants’ rights that, he explained, were recently distributed to immigrant communities and made to look as though they came from the ACLU, where Wizner directs the Speech, Privacy, and Technology Project. The deepfakes “frame,” he noted, is “much too small.”
The events of the last month have prompted me to reflect often on Wizner’s words. The spread of a global coronavirus contagion along with a concomitant “infodemic” of viral misinformation about the disease, its origins, and potential remedies, have made clear that disinformation can infect almost any facet of contemporary life. Human patterns of producing, sharing, and consuming information have entered a new era. And so too must our understanding of the dangers and opportunities presented by those changes.
To focus, as the public often has, only on a narrow-band of charismatic examples of disinformation—“fake news” about electoral candidates or highly realistic deepfake videos of celebrities or politicians, the most convincing of which are still largely in the future—is to “see through a glass, darkly”: understanding these issues only partially and incompletely.
NYU’s conference, co-sponsored by the NYU Journal of Legislation & Public Policy and the NYU Center for Cybersecurity, succeeded particularly well in countering this kind of myopia. The speakers and panelists covered the waterfront of issues and possible solutions.
The Deep Trust Alliance’s Kathryn Harrison presented the keynote address on the “coalition of stakeholders” from the government, private sector, and civil society that are needed to counter the dark side of deepfakes. She highlighted the wide berth of these challenges, including: the liar’s dividend (bad actors who challenge even true images as false), the manipulation of markets, false impersonation, and extortion and harassment.
The first panel, moderated by NYU Distinguished Fellow Judith Germano, focused on how technology companies, human rights organizations, and journalists are approaching these issues. One interesting issue the panel discussed was how disinformation that is narrowly tailored to small audiences—the electorate for a local school board election, for instance—could have disturbing effects, especially when it can be produced at scale. The speakers also noted how disinformation can be used to discredit journalists and activists, especially in countries that are politically unstable.
I was privileged to moderate a fireside chat with former American diplomats Emerita Torres now of the Soufan Center and Mounir Ibrahim now at Truepic on the implications of deepfakes on terrorism and other global challenges. Ibrahim spoke from his personal experience of how the liar’s dividend worked to stall intervention in Syria; the Damascus authorities just claimed that incriminating images were fake. Torres pointed out that that “[w]e need a multi-stakeholder, comprehensive approach” to these issues. She also noted we should develop education and public literacy around information and will need to build societal resilience.
The final panel, moderated by Randy Milch, the Co-Chair of the Center for Cybersecurity, focused on legislative and public responses to deepfakes, and explored the tension between unfettered speech and wise restraints. Lindsay P. Gorman of the Alliance for Securing Democracy made the apt observation that there may be more corporate buy-in to regulations if nonconsensual, deepfake pornography makes its way to mainstream platforms and negatively affects advertising. She also pointed to recent House of Representatives Ethics Guidelines against spreading deepfakes.
These House Guidelines are exactly the kind of accounting mechanism, which fall short of actual laws, that I believe can discourage the malicious uses of such technologies. There is a huge delta between doing nothing to address the dangers of disinformation and deepfakes and novel federal or state bills that actually become law. Ethical guidelines, professional rules, formal administrative regulations and informal policy guidance, and even social norms can serve important salutary functions. But to craft them well, we need to see the challenges of disinformation in the proper, broad context.
In an attempt to widen the framework for how society conceptualizes the dangers of disinformation, on March 12, 2020, I published with two colleagues a Client Alert, Identifying the Legal and Business Risks of Disinformation and Deepfakes: What Every Business Needs to Know (PDF).
Like the conference, this paper is meant as a tour of the horizon for private sector entities facing not just deepfakes but disinformation, more generally. The Alert:
- (1) reviews the business and legal risks that could arise when disinformation targets businesses, markets, and corporate leaders;
- (2) addresses potential causes of action that could be brought by victims of disinformation against the individual creators or propagators of malicious content, whether under long-established state and federal laws or under new laws enacted within the past year that specifically address deepfakes; and
- (3) reviews some best practices businesses can undertake today to prepare.
It also points out that this is not just an online problem but can pose risks offline, from social engineering to credential theft, from insurance fraud to falsified court evidence, and beyond.
It is my hope that this paper will contribute to the kind of expanded discourse the NYU conference encouraged—one that places in true light the potential benefits and dangers of, and possible solutions for, a technology that is so often used to spread deception.
Suggested Citation: Matthew F. Ferraro, Through a Straw Darkly: Reflections on the NYU Conference “When Seeing Isn’t Believing: Deepfakes and the Law”, N.Y.U. J. Legis. & Pub. Pol’y Quorum (2020).