Wave of new US bills seeks federal AI department and tougher oversight

In late September, US lawmakers introduced a flurry of federal bills aimed at tightening oversight of artificial intelligence. The most eye‑catching proposal in both the House and the Senate would create a dedicated Department of Artificial Intelligence, a cabinet‑level agency tasked with regulating advanced AI systems, including so‑called frontier models. The goal is to consolidate authority that is currently scattered across multiple departments.

Several companion bills focus on more specific safeguards. One would require all AI systems to include a human‑controlled shutdown mechanism, effectively mandating a “kill switch.” Another would expand federal data collection to track AI’s impact on the labor market, while additional measures push for broader transparency obligations on AI developers and deployers. All of these bills were referred to the relevant committees by September 24, setting the stage for hearings and potential mark‑ups.

Taken together, the proposals mark a shift from the current patchwork of executive orders and state‑level rules toward a more unified federal framework. Yet industry lobbying and partisan divides mean that the final shape of any legislation remains highly uncertain, and debates over how far to regulate frontier AI models are only just beginning.