The Trump administration is weighing a new approach to artificial intelligence oversight that would create an independent body to review the safety of leading AI models. The idea, described as similar in concept to the regulatory structure used in finance, would bring outside supervision to some of the most advanced systems.
According to the report, the proposed watchdog would evaluate model safety with input from the industry. That could mark a shift from more improvised government actions that companies say have created uncertainty around when and how powerful AI systems can be released.
The discussion appears to come after complaints from Silicon Valley leaders about the ad-hoc nature of recent US efforts to slow the rollout of cutting-edge models. A more formal review process could give developers clearer expectations while still addressing concerns about the risks posed by increasingly capable AI.
While the plan is still under consideration, it highlights the broader debate in Washington over how to regulate frontier AI. Policymakers are trying to balance innovation, competition and safety as advanced models become more central to business, research and public life.
