Anthropic's Project Glasswing: Rivals Unite to Tame AI's Hacking Prowess
Anthropic isn't releasing its powerful new Claude Mythos Preview to the world just yet. Instead, it's handing it to rivals like Microsoft and Google in Project Glasswing to probe for AI-driven cyber doom.
⚡ Key Takeaways
- Project Glasswing unites rivals like Microsoft and Google to test Anthropic's Claude Mythos Preview's hacking capabilities privately. 𝕏
- Mythos excels at exploit chains and binary analysis as a byproduct of code training, threatening core security assumptions. 𝕏
- This could birth new standards, echoing early internet defense pacts like CERT amid AI's dual-use cyber risks. 𝕏
Worth sharing?
Get the best Legal Tech stories of the week in your inbox — no noise, no spam.
Originally reported by Wired - AI
Related Stories
AI Regulation
Anthropic's Mythos Model Finds Thousands of Zero-Days—But the Real Risk Is Political
Governance & Ethics
Anthropic's Mythos: Cybersecurity Pretext or Enterprise Cash Grab?
Governance & Ethics
Black Forest Labs: How a 70-Person German Crew Outfoxes AI Image Giants
Governance & Ethics