Only five days! Anthropic's confidential Mythos model breaks through Apple's five-year M5 memory defense

AIMPACT News, May 16 (UTC+8), according to Beating Monitoring by Dongcha, the security research team Calif announced that they used the Mythos Preview model, which has not yet been publicly released by Anthropic, to successfully build the first publicly available macOS kernel memory corruption exploit chain on a device equipped with an M5 chip. MIE (Memory Integrity Enforcement) is a flagship hardware security mechanism developed by Apple for the M5 and A19 chips, with the company investing five years and billions of dollars. Its original design goal was not to be completely immune to hackers but to significantly increase the cost of exploitation to mitigate memory corruption vulnerabilities. The Calif team took only five days from discovering the flaw to completing the exploit. This attack chain involves two vulnerabilities and multiple techniques, starting from a non-privileged local user, relying solely on standard system calls, ultimately gaining root access to the device. The exploit belongs to pure data-driven kernel local privilege escalation, targeting macOS 26.4.1 running on real hardware with the kernel MIE mechanism enabled. Mythos models excel at quickly generalizing from learning a specific type of attack to similar issues, helping the team rapidly identify flaws belonging to known vulnerability classes, with subsequent efforts by human experts to tackle new hardware defenses. This breach validation demonstrates the high efficiency of the “AI discovering vulnerabilities + experts bypassing defenses” combination, and also proves that with the assistance of top-tier large models, small security teams can challenge the technical barriers built by large companies at great expense. (Source: BlockBeats)