Anthropic adds 150 partners to Claude Mythos amid flaw surge

Anthropic has expanded access to its AI model, Claude Mythos, to 150 additional organizations, enhancing its Project Glasswing initiative aimed at uncovering vulnerabilities in critical software. Originally limited to about 50 partners, the new organizations span over 15 countries and include sectors like healthcare and power, indicating their code's high potential impact in the event of cyber-attacks.

While over 10,000 critical flaws have already been identified using this AI, a growing concern remains: the discovery of vulnerabilities is outpacing efforts to resolve them, leading to increased pressure on security teams. Experts emphasize the importance of prioritizing and fixing issues swiftly to improve overall software safety.