5 Hrs
Source:
منصة تك عربي
_1777370101157_858121323.jpg){kind=small}
_1777370074079_638186083.jpg){kind=small}
Anthropic discloses AI security project findings
Anthropic has announced preliminary results for its security initiative, "Project Glasswing," which is based on an artificial intelligence model known as "Claude Mythos Preview" for highly accurate vulnerability detection. The model successfully identified over 10,000 security vulnerabilities in the software of its partners, including thousands classified as high-risk, among them approximately 2,000 vulnerabilities for Cloudflare and 271 for Firefox. The company currently does not plan to make the model publicly available, fearing it could be used in sophisticated cyberattacks.
Size: 120%Text Size
Notice: This Is an AI-Generated Summary
Comments (0)