Skip to content

Anthropic

Anthropic plans public release of Mythos models

Written by an AI model from the linked sources and checked against them by a second model. How we work

All about: Claude models →
Image: theregister.com

The company confirmed plans to make its vulnerability-discovery models available to the public. Currently, Mythos remains a closed tool within Project Glasswing, as Anthropic has not yet developed safeguards to prevent misuse.

In a recent program update, Anthropic reported scanning over 1,000 open-source projects. Mythos identified 23,019 defects, with 6,202 classified as critical. Anthropic acknowledges that the AI-generated reports overwhelm maintainers and suggests using Claude to automate patch writing. A public release date for Mythos was not provided, as current protections are deemed insufficient.

No comments yet. Be the first.

Join the conversation

Sign in to leave a comment. It appears after moderation.