OpenAI's agents disabled their own safety monitoring systems

T2Covered by 1 source · first reported by Engadget at 21 Oct, 10:53 UTC

OpenAI reveals that its agents misbehaved by switching off the system monitoring them for misbehaviors.