Anthropic plans to make its Mythos-level AI models available to all customers within weeks, following development of stronger safety measures. The company previously restricted access to the technology, citing security concerns.
Anthropic is moving forward with a broad release of AI models comparable to Mythos, technology the company had deemed too risky for public access. The San Francisco-based AI firm says the new models will be available to all customers "in the coming weeks" after implementing enhanced safeguards.
The announcement comes as Anthropic released Claude Opus 4.8 on Thursday, an upgraded version of its flagship model featuring improved coding and knowledge work capabilities. The new version maintains pricing at the same level as its predecessor.
Mythos-class models represent a significant capability tier in Anthropic's product lineup, with cybersecurity features on par with the original Mythos system. The company's previous hesitation to release these models reflected broader industry caution around deploying advanced AI systems without adequate safety measures in place.
Anthropric's decision to proceed with the wider rollout signals confidence in its safety infrastructure. The development of stronger safeguards—specific details of which remain undisclosed—apparently addressed the company's initial concerns about unrestricted access.
The release strategy aligns with Anthropic's competitive positioning against other major AI labs, including OpenAI and Google DeepMind. As the AI market intensifies, companies face pressure to make powerful models accessible while managing associated risks.
Anthropic has emphasized responsible AI development throughout its operations, including publishing research on AI safety and implementing various protective measures across its product line. The Mythos release represents a test of whether those safeguards can scale to broader user bases.
The timing of the announcement—during an accelerating period of AI model releases and upgrades—underscores the rapid pace of development in the sector. Anthropic's willingness to move from restricted to public access suggests the company believes its risk mitigation strategies have matured sufficiently for general deployment.
Prime Minister Anthony Albanese will frame artificial intelligence as a pivotal societal moment comparable to the renewable energy transition in a speech this week. The address will address AI safety concerns but will not include updates on copyright protections for artists.
Finnish AI lab NestAI Oy is deploying an AI engine for the Finnish and Estonian militaries to build independent models for military operations and autonomous systems.
A coalition of 200 economists and AI leaders has issued a stark warning about artificial intelligence's impact on employment. The group signals consensus that significant disruption to the labor market is coming.
Apple has released the first public betas of iOS 27, iPadOS 27, macOS 27 Golden Gate, watchOS 27, and tvOS 27. The rollout marks the public debut of Apple's redesigned Siri AI across its entire ecosystem.