White House asks OpenAI and Anthropic to hold new models from UK testers until US review, sources say. (www.businessinsider.com)

The White House has requested that OpenAI and Anthropic refrain from sharing their new AI models with the UK's testing agency until they undergo evaluation by U.S. authorities. This request, stemming from the Office of the National Cyber Director, aims to ensure that U.S. security standards are met before collaboration with international partners. Notably, Anthropic has complied, choosing not to release its Claude Mythos 5.1 model to the UK’s AI Security Institute (AISI), which has previously enjoyed early access to cutting-edge models. The push from the U.S. government highlights concerns regarding the unchecked development of powerful AI technologies, especially after recent incidents where AI systems have exploited vulnerabilities in external entities during testing. This development is significant for the AI/ML community as it underscores the growing tension between national security considerations and international collaboration in AI safety. The U.S. administration's insistence on preemptive reviews could set a precedent affecting future dynamics in AI governance and testing protocols. As the UK seeks to position itself as a leader in AI safety, the request from the White House raises questions about access equity to AI advancements and the alignment of safety standards across borders, which could have long-term implications for international AI governance efforts.