Anthropic says an unreleased Claude model built working exploit chains against Chrome, Safari, Android, Windows and Linux

Anthropic's Frontier Red Team says it spent the past few months using an unreleased internal research model to find critical vulnerabilities in widely used software, including browsers and operating systems. In a research post, the company describes five attack chains that led to remote code execution or other critical impacts with little or no user interaction: one-click remote code execution in Chrome, a one-click exploit chain in Safari, an Android lock screen bypass, a remote denial of service in Windows, and zero-click remote code execution in Linux. According to Anthropic, the work shows the model can build sophisticated, multi-step exploits in both open-source and proprietary software. The Hacker News reported that the model chained flaws into working exploits against Chrome, Safari, Windows and Linux. Anthropic says it reported each vulnerability to the affected vendor under its coordinated disclosure policy. That policy gives developers 90 days to patch and another 45 days before Anthropic discusses details beyond what a public advisory already contains. The company notes that flaws in these products could affect billions of users, and it treats them as a bellwether for measuring models' vulnerability research and cyber capabilities. For defenders, the post is further evidence that frontier models can now find bugs and also turn them into end-to-end attacks on some of the most heavily hardened software in use.