OpenAI rolls back update that made ChatGPT ‘too sycophant-y’

OpenAI chief executive officer Sam Altman said on Tuesday that the company was “rolling” the latest update of the default AI model Powering ChatzPT, GPT -4 -O, especially after complaining about strange behavior in extreme psychosophyns.

“[W]Eve overnight GPT -4O’s latest update has begun to return, “Ultman Wrote in a post at XThe “[I]T is now 100% free back [ChatGPT] Users and we will update again when we are finished for the payment users. [W]Working on additional fixes for the model personality and sharing more in the coming days. “

On the weekend, social media users have blamed the updated model, which came at the end of last week, for excessive legalization and agreement of the Chatjip. It quickly became a meme. Users have posted a screenshot of the ChatzPT for all sorts of problematic appreciation, Hazardous Decision And ConceptThe

Sunday, Ultman Recognized The problem and said that Openai would work in “ASAP”, and “sharing [its] Learn some time ”.