Yhn BusinessRetail first seen 13 h ago, last 52 min ago, peak #7
Anthropic says Chinese AI model has Mythos-class hacking abilities
Original: Anthropic claims popular Chinese AI model has Mythos-class hacking abilities
Anthropic's frontier red-teaming report claims a popular open-weight Chinese AI model shows what it calls Mythos-class hacking capabilities, alongside weak safeguards. The findings are drawing attention to the risks of freely downloadable frontier models, with debate over whether the security assessment holds up and how open-weight releases should be governed.
Why now: Anthropic's report spotlights security risks in a widely used open-weight Chinese AI model, sparking debate over AI safety and open-source policy.
Evidence
API: https://socialmediatrends-api.osmike.com/v1/trends/885405