Anthropic's Mythos AI helped find and exploit two unknown macOS kernel vulnerabilities in five days

Anthropic的Mythos AI在五天内协助发现并利用两个未知macOS内核漏洞

Anthropic’s Mythos AI helped researchers find two previously unknown macOS kernel vulnerabilities in five days and chain them into a privilege-escalation exploit that bypassed Apple’s memory integrity protection, according to the Wall Street Journal snippet.

Why it matters: HKR-H/K/R all pass, and Anthropic-linked AI security work is high-signal. The score stays in 78–84 because the source is a social post and lacks paper details, reproducible conditions, or exploit mechanics.