Anthropic, a leading Artificial intelligence firm, has launched the latest cybersecurity initiative, Project Glasswig, powered by its cutting-edge model, Claude Mytos Preview.
The recently introduced model is particularly designed to detect and resolve critical software vulnerabilities.
As per Anthropic, Claude Mythos has successfully detected thousands of high-severity zero-day vulnerabilities across major operating systems and web browsers, such as long-standing flaws in OpenBSD and FFmpeg.
Additionally, the Claude AI manufacturer announced its partnership with the leading tech players of today’s era, including Amazon Web Services, Apple, Microsoft, and Google, aiming to strengthen global digital infrastructure.
During experimentation, the AI demonstrated cutting-edge capabilities, including autonomously chaining multiple exploits and bypassing sandbox restrictions.
To prevent Mythos’ potential misuse, Anthropic has limited public access to the model. Rather, it provides restricted access to trusted organizations while committing up to $100 million in usage credits to improve defensive cybersecurity efforts.
Experts raised concerns regarding such advancements, as they could speed up both cyber defense and cyberattacks. As per industry leaders, tasks that once used to take longer time can now easily be completed in minutes using AI.
Anthropic underscored that these excellent capabilities emerged naturally from coding enhancements and reasoning, indicating the latest phase in cybersecurity where AI could redefine both the threat and protection realm.
