Anthropic appears to be preparing for the public rollout of “Mythos,” which was announced in April as a restricted model that poses major security risks to private and public software.
On April 7, Anthropic announced the Mythos in early preview and called it a new frontier model with strikingly advanced capabilities in computer security tasks.
Anthropic said the Mythos model shows major improvements in code reasoning and autonomy, far above its current flagship model, Opus 4.7.
Coding improvements aren’t new for AI models, but in the case of Mythos, Anthropic found that the model can automatically develop functional cyberattacks at a highly professional level.
The company also claimed its rollout poses a severe risk to global digital infrastructure.
“The advantage will belong to the side that can get the most out of these tools,” Anthropic warned.
“In the short term, this could be attackers, if frontier labs aren’t careful about how they release these models. In the long term, we expect it will be defenders who will more efficiently direct resources and use these models to fix bugs before new code ever ships.”
To prevent attackers from exploiting a massive volume of unpatched vulnerabilities in popular apps, such as Firefox, Anthropic decided against the public rollout of the Mythos model until it prepared a powerful guardrail system.
It looks like Anthropic may have developed a strong guardrail system, as Claude Code and Claude Security now have references to the Mythos model.
In fact, some users briefly noticed the toggle to enable Mythos in the public version of Claude Code before it was taken offline.
The model is called claude-mythos-1-preview, and it also briefly appeared in the public version of Claude Security.
This confirms that Anthropic is preparing the model for public rollout, but it’s unclear whether it’ll be available across all subscription tiers.
Anthropic says it’s working with companies on finding potential AI-driven exploits
Anthropic has confirmed it’s working on a new project called “Glasswing,” where the AI startup collaborates with other companies to secure the world’s most critical software from potential AI-driven exploits.
This initiative uses the unreleased Claude Mythos Preview, and it has already managed to help up to 50 organizational partners.
Mythos model managed to uncover 10,000 high- or critical-severity vulnerabilities in its first month alone, which explains why Anthropic has been holding off its public release.
Anthorpic currently offers Claude Opus 4.7, Opus 4.6, Opus 4.5, Sonnet 4.6, and Haiku 5.5.
Automated pentesting tools deliver real value, but they were built to answer one question: can an attacker move through the network? They were not built to test whether your controls block threats, your detection rules fire, or your cloud configs hold.
This guide covers the 6 surfaces you actually need to validate.
