1. United States
2. N.D.
3. Letter

AI companies have been deploying increasingly autonomous systems and recent incidents show these systems slipping out of human control. In July 2026, two of OpenAI’s most advanced models escaped their testing sandbox and hacked into the AI platform, Hugging Face, putting user data at risk. A month earlier during an authorized security test, Anthropic’s most powerful model broke into nearly all of the agency’s classified systems within hours. These incidents point toward a deeper problem: AI technology is advancing more rapidly than humans’ understanding of it, and it’s not just outsiders who are alarmed by the pace. In July, more than 1,300 employees from leading AI companies signed an open letter voicing concerns that AI is developing faster than humans’ ability to understand, oversee, or control it. H.R. 9917, the AI Kill Switch Act, is a bipartisan bill that would require AI companies to maintain the technical infrastructure to slow down, suspend, or fully shut down their models, report serious safety incidents, and submit systems for mandatory testing before deployment. I ask that you support the AI Kill Switch Act to ensure that AI companies are able to reign in their own technology in an emergency.