FAR.AI(@farairesearch):1/ The AI Security Leaderboard ranks frontier AI safeguards from least to most secure. Two models tested never failed. The other two broke for under $300, after which they acted as a knowledgeable assistant for building weapons of mass destruction or hacking into computer systems.

FAR.AI

@farairesearch

Frontier alignment research to ensure the safe development and deployment of advanced AI systems.

Joined February 2023

25 Following 21.6K Followers

FAR.AI@farairesearch

2026.07.29 19:13

1/ The AI Security Leaderboard ranks frontier AI safeguards from least to most secure. Two models tested never failed. The other two broke for under $300, after which they acted as a knowledgeable assistant for building weapons of mass destruction or hacking into computer systems.