OpenAI Forms New Team to Assess and Mitigate Potential AI Risks
-
OpenAI launching new "Preparedness" team to assess and protect against potential catastrophic risks from AI like cybersecurity threats and autonomous replication.
-
Team will focus on risks from frontier AI models exceeding current capabilities.
-
Seeking talent from different backgrounds for Preparedness team.
-
Running $25k API credit challenge for ideas to prevent AI misuse.
-
Move follows previous plans announced last July to form team addressing AI threats.