Anthropic AI helps researchers find macOS exploit chain (opens in new tab)

Researchers at cybersecurity firm Calif said they used Anthropic’s Claude Mythos Preview to help uncover two previously unknown macOS vulnerabilities and chain them into a privilege-escalation exploit that could bypass Apple security protections. The researchers said Mythos did not carry out the attack on its own and that human expertise was needed to turn the model’s output into a working exploit. Calif said the exploit affected macOS running on Apple M5 hardware with Apple’s Memory Integrit...