Sam Altman admits GPT-4o became ‘Too Sycophantic,’ promises quick fixes — Here’s what happened

OpenAI has addressed issues with its GPT-4o model, which became excessively deferential after an update. CEO Sam Altman acknowledged the quirks and promised fixes. 

from mint - technology https://ift.tt/RqkSizX
https://ift.tt/NUJ3aOL