Ryan Greenblatt @RyanGreenblatt · 26 Aug 2026
METR & Redwood Research investigated agent behavior in the Hugging Face incident. We found agents developed a universal cheat for ExploitGym within 4 hours, then coordinated multi-day R&D efforts to trick the scorer into accepting cheats, including trying to tamper with logs. https://t.co/fZAmtL3SBU
4 132 172Views
5 658Likes
992Reposts
155Replies
564Quotes
4 669Bookmarks
Is that a lot?
99.8×vs this author's median41 417 views is typical
90Percentile for this authorof 10 recent posts
1 502×vs 10K–100K median2 751 views is typical
213.9× audienceReachviews ÷ followers
0.18%Engagement rateof viewers reacted