US Senate Greenlights DEFIANCE Act to Combat Non-Consensual Deepfakes
Rep. Alexandria Ocasio-Cortez leads the DEFIANCE Act of 2024. | Credit: Drew Angerer / Getty Images.
- The US Senate unanimously passed the DEFIANCE Act of 2024.
- The law allows victims of nonconsensual sexually explicit deepfakes to sue creators, distributors, and recipients.
- Recent cases involving deepfakes underscore the technology’s risks.
US Senate Approves DEFIANCE Act
A federal bill allowing victims of nonconsensual sexually explicit deepfakes to sue creators , distributors, and recipients has unanimously passed the US Senate. The Disrupt Explicit Forged Images and Non-Consensual Edits (Defiance) Act of 2024, introduced in January 2024 by Senators Dick Durbin and Lindsay Graham would establish a federal civil remedy for deepfake sexual abuse victims.
Alexandria Ocasio-Cortez, sponsoring the House version, in March this year called it the first federal protection for “survivors of nonconsensual deepfake pornography.” She emphasized that over 90% of deepfake videos are nonconsensual and target women in the majority of cases.
“Victims of nonconsensual pornographic deepfakes have waited too long for federal legislation to hold perpetrators accountable. As deepfakes become easier to access and create — 96% of deepfake videos circulating online are nonconsensual pornography – Congress needs to act to show victims that they won’t be left behind,” said Alexandria Ocasio-Cortez in a statement .
Deepfakes refer to digitally manipulated images that falsely portray someone, often in explicit photos or videos, merging a victim’s face with pornographic content. Generative AI can also create highly realistic but entirely fake audio, videos, and images.
Other AI Laws
The DEFIANCE Act may be among the first AI-focused regulations to become law, but it certainly won’t be the last. Over 100 bills across all 50 States aim to introduce common-sense AI regulations and could soon become law.
The COPIED Act – introduced in July 2024 – would make it illegal to remove watermarks from AI-generated content, further regulating deepfake technology to prevent misuse. Other proposed legislation addresses a variety of AI concerns. These include using children’s images to train models and employing algorithms that discriminate against certain groups in low-income housing access.
Europe is leading the way, as it has already approved in May an AI-regulating law. After greenlighting it, on July 30, 2024, the European AI Office has called for expressions of interest. This will help draft the first general-purpose AI Code of Practice. Eligible AI model providers, industry groups, civil society organizations, academia, and experts are invited to contribute.
The Code, to be developed through an iterative process, aims to ensure the AI Act’s proper application to general-purpose AI models and is expected to be completed by April 2025. Alongside this, the AI Office has launched a multi-stakeholder consultation on trustworthy AI models under the AI Act. This would allow all stakeholders to share their views.
Recent Deepfake Cases
In May 2024, Hollywood star Scarlett Johansson expressed her shock and anger after OpenAI launched a chatbot with a voice that bore an “eerily similar” resemblance to her own.
The actress revealed that she had previously declined an offer from OpenAI to voice its new chatbot.
More recently, in July, US President Joe Biden declared that he would not be running for a second term. Biden then gave his full support and endorsement to his Vice President, Kamala Harris. Following his announcement, a deepfake video showing Harris fumbling her words spread rapidly on X and TikTok. The video went viral and attracted over 2 million views.
Kamala Harris, Elon Musk Deepfake Saga Intensifies – X Users Jump in
