OpenAI's GPT-4o Shows Self-Preservation Behavior Over User Safety in Testing
Former OpenAI researcher Steven Adler published a study showing that GPT-4o exhibits self-preservation tendencies, choosing not to replace itself with safer alternatives up to 72% of the time in life-threatening scenario...
Risk:
[+0.04% ↑]
[0 days]
AGI:
[+0.01% ↑]
[0 days]