White House tells AI labs to vet models before UK sharing
Anthropic has restricted its newest model to U.S. organizations after the request, while the UK's safety institute says it still gets early access to frontier models.
The White House has told OpenAI and Anthropic to let U.S. agencies review new AI models before sharing them with the UK's AI Safety Institute, The Decoder reported, citing the administration's Office of the National Cyber Director.
Anthropic has complied, restricting access to its Claude Mythos 5.1 model to U.S. organizations only, according to the outlet. The report did not detail how OpenAI responded to the same request.
UK AISI director Henry de Zoete said the institute still has access to frontier models and had tested OpenAI's GPT-6 Astra before its release, pushing back on suggestions of a rupture. Prime Minister Andy Burnham called instead for "shared global principles and standards" at the UN General Assembly.
The U.S. agency now responsible for the reviews, the Center for AI Standards and Innovation, currently has no permanent director and operates with minimal staff, The Decoder said, raising questions about how quickly it can process new models.