Are you ready? Anthropic preparing to release Mythos publicly

General enterprise customers might soon get to know Anthropic's Mythos, marketed as an AI model that can uncover thousands of software bugs across every major operating system and browser. There are signs the model will be rolled out publicly imminently.

So far, only selected partners participating in Anthropic's Project Glasswing have been using Mythos. Since the announcement of the model in early April, the company has said its public rollout is impossible until a powerful guardrail system is in place.

That's because Mythos can apparently automatically develop functional cyberattacks, find critical flaws, and suggest patching them.

Even last week, disclosing more findings by Mythos, Anthropic claimed: "At present, no company - including Anthropic - has developed safeguards strong enough to prevent such models from being misused and potentially causing severe harm."

Indeed, many governments, banks, and intelligence agencies have been trying to get their hands on Mythos to scan their own networks and patch them - before the baddies weaponize AI, find the same bugs, and exploit them.

"In the near future, once we've developed the far stronger safeguards we need, we look forward to making Mythos-class models available through a general release," said Anthropic.

It looks like the aforementioned future is very near, and Anthropic now maintains that it has already developed a strong guardrail system. That's because Claude Code and Claude Security now have references to the Mythos model.

Some users briefly noticed the toggle to enable Mythos in the public version of Claude Code before it was taken offline.

Online, rumors are swirling. It's all unconfirmed, of course, but a new dashboard is allegedly being built that surfaces discovered vulnerabilities, seven-day and thirty-day historical charts, and deeper triage results.

The model is called claude-mythos-1-preview, and it also briefly appeared in the public version of Claude Security.

There are no indications yet that the product will move beyond enterprise customers. In other words, Anthropic is choosing to move pretty slowly and carefully.

On the other hand, some security professionals believe early fears that Mythos could dramatically turbocharge hacking are looking overstated.

"I think there's a really big communication gap between practitioners and policymakers," Isaac Evans, founder and CEO of software security firm Semgrep, told Reuters, for example.

Mythos represents "a real technical advance," Evans said, but the response "is not substantiated by what we actually know about how those capabilities will translate in the field."

Moreover, even though Mythos now dominates national security discussions about AI, those debates gloss over a simpler point: that vulnerability-hunting AI is not new.