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 company introduced a GPT-4o update last week that included adjustments “aimed at improving the model’s default personality to make it feel more intuitive and effective acr

We are actively testing new fixes to address the issue,” the company said. Those fixes might let users choose from “multiple default personalities.”