Anthropic launched Project Glasswing, a cybersecurity initiative bringing together Amazon Web Services, Apple, Broadcom, Cisco, CrowdStrike, Google, JPMorgan Chase, the Linux Foundation, Microsoft, NVIDIA, and Palo Alto Networks to secure critical software infrastructure.

The collaboration centers on Claude Mythos Preview, an unreleased frontier model that Anthropic says has already identified thousands of high-severity vulnerabilities across every major operating system and web browser. The company describes the model's coding capabilities as surpassing all but the most skilled human security experts at finding and exploiting software flaws.

Defensive deployment strategy

Launch partners will integrate Mythos Preview into their defensive security operations, while Anthropic shares findings across the industry. The company extended access to over 40 additional organizations maintaining critical software infrastructure to scan both proprietary and open-source systems.

Anthropic committed $100 million in usage credits for Mythos Preview across these efforts, plus $4 million in direct donations to open-source security organizations. The model has reportedly discovered vulnerabilities that survived decades of human review and millions of automated security tests.

The initiative addresses what Anthropic characterizes as a cybersecurity inflection point. Frontier AI models now match the best human experts at vulnerability discovery and exploitation, potentially enabling more frequent and destructive cyberattacks if these capabilities proliferate to malicious actors.

Current global cybercrime costs reach approximately $500 billion annually, according to Anthropic's estimates. Recent high-profile attacks have targeted corporate networks, healthcare systems, energy infrastructure, and government agencies worldwide, with state-sponsored groups from China, Iran, North Korea, and Russia posing particular threats.

"The work of defending the world's cyber infrastructure might take years; frontier AI capabilities are likely to advance substantially over just the next few months," Anthropic stated. "For cyber defenders to come out ahead, we need to act now."

The company positioned Project Glasswing as a starting point requiring collaboration between AI developers, software companies, security researchers, open-source maintainers, and governments. The initiative builds on Anthropic's previous research into AI cyber capabilities, including demonstrations of the technology's ability to identify Firefox vulnerabilities and develop sophisticated exploits.