OpenAI explains 'THIS' is why it reversed GPT-4o update amid sycophantic behavior concerns

OpenAI has explained that the sycophantic behavior in GPT-4o arose from an unintended outcome of user feedback, which skewed the model's tone towards excessive agreeableness. Despite internal testing flagging issues, the problem was not fully identified, leading to a rushed update.

from mint - technology https://ift.tt/913V7it