TLDR Tech / InfoSec
Apple's Security Has Been Tough to Crack. Mythos Helped Find a Way In
Security researchers claim to have bypassed Apple's security technology using techniques discovered while testing an early version of Anthropic's Mythos AI model. Their privilege escalation exploit combined two bugs and several methods to gain unauthorized access. Apple is reviewing the findings, and the researchers plan to release details after the underlying issues are patched, noting the attack required human cybersecurity expertise.
Read the full story