White House Wants First Look. UK Gets Seconds. Sovereignty Trumps Cooperation.
The White House has asked OpenAI and Anthropic to withhold new AI models from the UK's AISI testing agency until US government review is complete. The UK AISI, created in 2023, has routinely tested models before public release on a voluntary basis. The US body reviewing models is reportedly operating without a permanent director and facing bandwidth problems.
This demonstrates jurisdictional friction in AI governance. When two regulatory bodies both want pre-deployment access to frontier models, the one with territorial authority over the company wins. The mechanism here is regulatory leverage: the US can compel cooperation from US-domiciled labs in ways the UK cannot. The lesson for any observer is that AI safety is not just a technical problem, it is a geopolitical one.
OpenAI and Anthropic are the labs named in the request. The UK AISI, established in 2023, and an unnamed US review body are the agencies involved, with the US body reportedly understaffed and lacking permanent leadership.
- Visit aisi.gov.uk to read about the UK AISI's stated mission and voluntary testing framework.
- Search for the US AI Safety Institute's public materials to compare its stated scope and current leadership status.
- Read either OpenAI or Anthropic's public safety policy page to see how each lab describes its own pre-release testing commitments.