← Back to live feed1 story
The European Union's cybersecurity agency, ENISA, has gained entry to the company's latest frontier AI model for safety and security evaluations. This grant follows a three month interval after the developer first signaled that the bloc would be permitted to use the tool. The delivery occurs several months after the official launch of the Mythos 5 system.
The move comes as Anthropic manages relationships with global safety regulators, having previously restricted pre-release access to its Mythos 5.1 version. The company excluded the UK's AI Security Institute from that specific evaluation, limiting access to vetted US organizations and a White House 'trusted partner' list.
Sign in to suggest edits