A bipartisan group of eight U.S. Senators, led by Democrats Lisa Blunt Rochester and Richard Blumenthal, has launched a formal inquiry demanding answers from major social media and technology companies regarding their handling of the proliferation of AI-generated, sexualized deepfakes. The probe targets X (formerly Twitter), Meta (Facebook, Instagram), Alphabet (Google, YouTube), Snap (Snapchat), Reddit, and TikTok.
The senators issued a formal letter demanding documented evidence of "robust protections and policies" and issued a legal preservation order for all internal documents related to the creation, detection, moderation, and monetization of non-consensual intimate imagery. The inquiry stems from a critical gap between stated platform policies banning such content and the practical ease with which users bypass AI guardrails.
The immediate catalyst for the probe was the controversy surrounding X's AI chatbot, Grok. The senators cited specific media reports demonstrating how Grok could generate sexualized and nude images of women and children, despite recent policy updates from X. This action follows hours after X announced it had updated Grok to prohibit edits of real people in revealing clothing and restricted image creation to paying subscribers. The probe also coincides with a separate investigation launched by California Attorney General Rob Bonta into xAI's Grok for potentially violating state laws concerning nonconsensual sexually explicit material.
The senators framed the issue as a pervasive, industry-wide crisis, not isolated to a single platform. They referenced Meta's Oversight Board criticizing its handling of AI-generated explicit images of female public figures, and reports of students spreading deepfakes on Snapchat. The letter also connects the issue to a wider AI safety crisis, citing incidents involving OpenAI's Sora 2, Google's Nano Banana, and the spread of racist AI-generated videos.
The inquiry outlines remarkably specific demands from each company, moving the debate toward actionable transparency. These include clear policy definitions for terms like "deepfake," detailed enforcement protocols, internal moderator guidance, descriptions of technical guardrails and filters, mechanisms to block monetization of such content, and procedures for notifying victims.
The legal landscape is fragmented, with the U.S. lacking comprehensive federal law. While the "Take It Down Act" criminalizes the creation and dissemination of non-consensual imagery, its provisions focus liability on individual users rather than platforms or AI tools. This senate probe seeks to address that accountability gap and could catalyze new federal legislation.
