Anthropic Launches Project Glasswing Using Mythos AI to Uncover Software Flaws

Anthropic unveiled a preview of its Mythos frontier model as the core of Project Glasswing.

“Project Glasswing Securing critical software for the AI era Introduction Today we’re announcing Project Glasswing1, a new initiative that brings together Amazon Web Services, Anthropic, Apple, Broadcom, Cisco, CrowdStrike, Google, JPMorganChase, the Linux Foundation, Microsoft, NVIDIA, and Palo Alto Networks in an effort to secure the world’s most critical software”

Anthropic

Mythos has identified thousands of zero-day vulnerabilities, some dating back decades.

Anthropic

Launch partners include Amazon, Apple, Google, Microsoft, Nvidia, and others.

Anthropic is committing $100 million in usage credits and $4 million in donations.

The model will not be generally released due to dual-use concerns.

Mythos’s capabilities reveal the paradox of advanced AI in cybersecurity.

The model can find bugs and also write proof-of-concept exploits.

CNBC

In testing, Mythos found bugs in every major operating system and web browser.

Anthropic’s red team lead said they did not feel comfortable releasing it generally.

Mythos found a 27-year-old bug in OpenBSD and a chain of vulnerabilities in Linux.

Project Glasswing represents an unprecedented pooling of competitors.

“Anthropic on Tuesday announced an advanced artificial intelligence model that will roll out to a select group of companies as part of a new cybersecurity initiative called Project Glasswing”

CNBC

Anthropic plans to share learnings so the entire industry can benefit.

Mythos was publicly discovered through a security slip-up weeks earlier.

Anthropic has engaged in ongoing discussions with federal officials.

Experts warn the AI arms race requires new public-private approaches.

Critics note risks that Mythos could yield new offensive vulnerabilities.