The fact
The institute will evaluate Anthropic and OpenAI models following the British AISI model to identify security risks.
The EU remains dependent on US and Chinese models, raising sovereignty concerns given ties between providers and their governments.
Click the link to read an article on the topic: