What happened
Anthropic expanded Project Glasswing on June 2. The program now reaches about 150 more organizations across over 15 countries. Glasswing is built around Claude Mythos, a frontier model that is strong at finding and fixing software vulnerabilities.
An earlier group of around 50 partners used Mythos Preview to find more than ten thousand high or critical severity vulnerabilities. Anthropic says Mythos can find weaknesses in major operating systems and web browsers, and chain them together.
Why it matters
AI that can find software flaws cuts both ways. In the right hands, it helps defenders fix problems before attackers find them. In the wrong hands, it could speed up attacks. Anthropic's careful rollout aims to give defenders a head start.
This is a new and fast growing market where AI meets cybersecurity.
MintedBrain take
AI in security is one of the most important and double-edged trends to watch. For organizations, tools that find flaws early are valuable, but they also raise the stakes if misused. Expect security to be a major AI battleground in the years ahead.
Discussion
Sign in to comment. Your account must be at least 1 day old.