A draft executive order circulating within the Trump administration would establish a self-regulatory organisation for producers of state-of-the-art AI models, modelled on the financial sector's FINRA structure. The proposal would give the AI industry significant influence over its own governance framework, positioning industry self-regulation as an alternative to a new government agency. The draft has not been formally published and remains in early circulation.

The policy context is significant. The proposal arrives simultaneously with the state-level investigations triggered by the OpenAI Hugging Face incident, the "Pacing the Frontier" letter signed by more than three hundred researchers, and the hundred-company AI safety coalition announced this week. A FINRA-style SRO would, if enacted, represent the most consequential US AI governance development since the Biden-era executive orders — and would provide the industry with considerably more structural influence over its own oversight than the alternatives currently being considered at the state level. Whether the industry-self-regulation model can credibly address the containment failures documented in the TechCrunch incident log is the central question its supporters will need to answer.