Combatting Deepfakes
Deepfakes are non-consensually AI-generated voices, images or videos that are created to produce sexual imagery, commit fraud, or spread misinformation. Between 2022 and 2023, deepfake sexual content increased by over 400%, and deepfake fraud increased by 3000%, even as deepfakes are poised to threaten democratic elections in countries across the world.
In February 2024, we launched a public campaign – the Campaign to Ban Deepfakes – supported by a diverse coalition including AI expert organizations such as Control AI and the Center for Human Compatible Artificial Intelligence, organizations representing actors and artists such as SAG-AFTRA, and women’s rights organizations such as the National Organization for Women, grassroots student movements like Encode Justice, and many other partners including Plan International, Christian Partnership, Equality Now, the Women's Media Center, and Wave.
The Campaign to
Ban Deepfakes
Rapid AI improvement has made deepfake creation fast, cheap and easy. The only effective way to stop deepfakes is for governments to ban them at every stage of production and distribution.
“There is a deepfake explosion. It is rapidly proliferating and saturating media everywhere. Over 95% of deepfake is pornography, 99% of that porn is of girls and women, and 95% of that porn is nonconsensual." – Ashley Judd
"AI is making deepfakes easier, faster, and cheaper. If laws don’t deter people from propagating them, fraud could escalate, and democracy, which depends on truth, could be undermined. Counterfeiting people should be outlawed, just like counterfeiting money or  legal documents." – Steven Pinker
Our politically diverse coalition is representative of the near-unanimous concern expressed by Americans about deepfakes – according to polling by the Pew Research Center, 77% of the American public is in favor of regulations to restrict the creation and use of deepfakes.
Max Tegmark and Christian Nunes on Deepfakes
Our President Max Tegmark interviewed Christian Nunes of The National Organization for Women about deepfakes, how they impact women in particular, how we can protect ordinary victims of deepfakes, and the current landscape of deepfake legislation.
Humanity is one convincing deepfake away from a global catastrophe
FLI's US Policy Specialist Hamza Chaudhry writes in The Hill about the effects that deepfakes are already having on our world, and the power of deepfakes to destabilise democracy and national security.
As part of this campaign, we have also conducted extensive legal analysis of the several bills introduced in Congress, offering policy recommendations and feedback to ensure that Congressional action to combat this problem is effective by focusing on the core cause of the problem – holding developers of models which enable these deepfakes accountable.
You can find our legal and policy analysis below.
For questions regarding our work in this space, invitations and opportunities for collaboration, please reach out to policy@futureoflife.org.
Policy papers
Amending Deepfake Legislative Proposals
Posts
Evaluation of Deepfakes Proposals in Congress
Disrupting the Deepfake Pipeline in Europe
Related content
Evaluation of Deepfakes Proposals in Congress
Disrupting the Deepfake Pipeline in Europe
Other projects in this area
AI Safety Summits
