AIThe Hacker Newsabout 1 hour ago
Anthropic AI Model Chains Browser Exploits
4 min read
Anthropic reports an unreleased AI model that chains five separate security flaws into working proof-of-concept exploits for Chrome, Safari, Windows, and Linux. The company stated the work required only initial prompting and little human involvement. While most flaws were patched before public release, the Linux chain allows root access over the network.
Level
Hype check
Tap to vote and see what everyone thinks.
