The article discusses new US legislative efforts regarding AI safety, specifically the introduction of the FRONTIER Act by Representatives Trahan and Obernolte, which federalizes existing state frameworks for model reporting and risk definitions. Concurrently, Senators Lieu and Moran introduced the AI Kill Switch Act to mandate shutdown capabilities for advanced models, while OpenAI CEO Sam Altman visited Washington to preview GPT-6 and discuss voluntary testing frameworks with the White House.
- The FRONTIER Act establishes an Under Secretary of Commerce for AI Security with emergency shutdown authority, though it lacks binding catastrophic risk reduction requirements and preempts state laws.
- The AI Kill Switch Act requires covered entities to shut down model inference upon request or crisis, with fines up to $20 million per day, but exempts open-weight models.
- Altman met with officials including Lutnick and Bessent, stating OpenAI supports mandatory pre-deployment testing and federal independent auditing while emphasizing the importance of pacing AI development.
- Nvidia's Jensen Huang advocated for open models to head off regulatory actions, though Senator Warner cautioned against a "pedal to the metal" approach without guardrails.
The piece highlights the ongoing tension between regulating frontier labs and preserving open-source ecosystems, noting that current bills struggle to balance safety requirements with the inherent risks of unrestricted model distribution.