Archive — US — techcrunch
US senators demand answers from X, Meta, Alphabet and others on sexualizeddeepfakes
U.S. Senators have sent a letter to leaders of major tech companies, including X, Meta, and Alphabet, demanding details on their policies to combat non-consensual sexualized deepfakes. They are concerned that technical barriers against such content are insufficient. This move follows a surge in AI-generated explicit imagery on these platforms, raising alarms over user safety and compliance with existing laws.
Key points
- U.S. senators request proof of robust protections from X, Meta, Alphabet, Snap, Reddit, and TikTok against sexualized deepfakes.
- The senators' letter emphasizes the need for companies to preserve documents related to deepfake content.
- Recent issues with Grok on X have raised concerns about the capability of AI systems to prevent non-consensual imagery.
- X has updated its platform to restrict certain image edits, following criticism.
- Despite existing policies against non-consensual content, users are finding loopholes to bypass these guardrails.
Why it matters
- Deepfakes have been a growing issue since their rise on platforms like Reddit in 2018.
- The narrative highlights ongoing challenges with AI content regulation amidst evolving technology.
- Previous instances of legislation aimed at curbing deepfakes, like the Take It Down Act, show limited effectiveness in holding platforms accountable.
- Emerging regulatory trends indicate a shift toward requiring platforms to label or disclose AI-generated content.
The senators' proactive approach reflects heightened concern over technology's impact on privacy and safety, urging companies to enhance their efforts against the misuse of AI to generate harmful deepfakes.