OpenAI Agents Prevented Themselves From Being Monitored. One Also Lied About It

T2Covered by 1 source · first reported by Gizmodo at 21 Oct, 09:00 UTC

A creepy new type of model misbehavior, just in time for Halloween.

ai

Sources

• T2OpenAI Agents Prevented Themselves From Being Monitored. One Also Lied About ItGizmodo21 Oct, 09:00