The recent Firefox 150 release includes fixes for 271 vulnerabilities identified using Claude Mythos Preview AI. Mozilla revealed the information in a new blog post, detailing how they’ve been using an early preview of the unreleased AI model as part of Project Glasswing.
Anthropic have been talking up their new Claude Mythos, which is apparently so powerful and dangerous at finding security vulnerabilities in various software that they’ve currently chosen to keep it private so only a few select companies and organisations have access to it in Project Glasswing.
We can expect to see much more of this appear as all these AI models expand and improve, even though some of them enable anyone to spam projects with completely junk reports – it seems Claude Mythos is actually helping identify some legitimate serious issues.
For all the hype and potential fear coming from it though, Mozilla do state that “Encouragingly, we also haven’t seen any bugs that couldn’t have been found by an elite human researcher” and they don’t believe that these AI models will find “entirely new forms of vulnerabilities that defy our current comprehension”.
It seems like every bit of major software will be using some form of AI for something, at least checking over for security issues feels like a good sane usage for it. Even the Linux kernel recently introduced documentation to allow AI code helpers, under certain conditions.
Article taken from GamingOnLinux.com.
Click Here For The Original Source.
