Anthropic pauses Mythos release, restricts access amid exploit concerns
Anthropic has put the brakes on releasing its latest AI model, Mythos, after concerns it could be used to find and exploit software bugs.
Designed to spot hidden system flaws and even create working exploits, Mythos was seen as both a breakthrough and a risk, so for now, access is restricted to a small group of partners, including Google, Microsoft, Amazon Web Services, NVIDIA, and JPMorgan Chase.
Mythos uncovers 27-year OpenBSD vulnerability
Mythos made headlines when it uncovered a 27-year-old vulnerability in OpenBSD, which is known for being super secure.
While this shows how game-changing the tech could be for cybersecurity, it also highlights why Anthropic is taking things slow.
The AI community is watching closely as the company tries to balance innovation with keeping systems safe.