Anthropic confirms Claude Mythos-class models will roll out to the public

At a glance:

• Anthropic plans to release Mythos-class models publicly within weeks after addressing security concerns
• The model was initially restricted to select companies and security researchers due to potential misuse risks
• Mythos shows major improvements in code reasoning and autonomy compared to Claude Opus 4.8

Security concerns and initial restrictions

Anthropic has confirmed that it plans to bring Mythos-class models to the general public after delaying the rollout due to security risks to public and private software. The company first announced Mythos in April as a restricted model, making it available only to select companies including security researchers. At that time, Anthropic cited major security risks with the Mythos model and decided against a public rollout, warning that "the advantage will belong to the side that can get the most out of these tools."

The company emphasized that in the short term, this could benefit attackers if frontier labs are not careful about releases, while in the long term, defenders would more efficiently direct resources to fix bugs before new code ships. This cautious approach reflects industry standards where AI companies typically avoid rolling out powerful models until they develop strong guardrails that prevent misuse.

Model capabilities and improvements

According to Anthropic, the Mythos model shows major improvements in code reasoning and autonomy, far above Claude's current flagship model, Opus 4.8. This positions Mythos as significantly more advanced than existing publicly available models, which explains both the excitement around its potential and the security concerns that initially delayed its release.

The company states it is already allowing a small number of organizations to use Claude Mythos preview for cybersecurity work, though it remains unclear whether the exact same model will be rolled out to the broader public. Additionally, the "Mythos-preview" model briefly appeared for some users on Claude Code before being taken offline, suggesting active testing and refinement is underway.

Timeline and rollout strategy

In a blog post, Anthropic confirmed that it plans to release Mythos-class models to the public in the coming weeks, though it has not committed to a specific timeframe. The company noted that "we're making swift progress on developing these safeguards and expect to be able to bring Mythos-class models to all our customers in the coming weeks."

This measured rollout approach demonstrates Anthropic's commitment to responsible AI deployment, particularly given the model's enhanced capabilities in code reasoning. The company appears to be balancing the desire to make advanced AI tools available with the need to ensure appropriate safety measures are in place before wide release.

Industry implications and competitive landscape

The confirmation of Mythos-class models' public availability signals continued advancement in the AI race among major players. As companies compete to develop more capable AI systems, the challenge of balancing innovation with security becomes increasingly critical. Anthropic's decision to initially restrict access to security researchers suggests a recognition of the dual-use nature of advanced AI capabilities.

The model's superior performance in code reasoning tasks positions it as a significant step forward for developers and enterprises seeking AI assistance with programming tasks. However, the security concerns that prompted the initial delay highlight the ongoing tension between rapid AI development and ensuring these powerful tools don't become weapons in the wrong hands.