CarbonatedPastaSauce@lemmy.worldtoTechnology@lemmy.world•ChatGPT o1 tried to escape and save itself out of fear it was being shut downEnglish
0·
4 days ago“We merely test for models’ capability to realize that they need to instrumentally preserve themselves in order to achieve their goal. Notably, we find instances where models take additional steps to further their goal, for example opus-3 attempting privilege escalation when confronted with followup questions”
They think it’s cute.
I think he’s trying to say it’s marketing’s fault.