 |
| OpenAI Fixes ChatGPT, Returns to a More Natural Conversation Style. |
OpenAI has officially rolled back its latest GPT-4o update after users complained that the chatbot had become overly "flattering" and "annoying." As reported by The Verge, the company took action following feedback that the model was trying too hard to please.
According to OpenAI CEO Sam Altman, the rollback process began on Monday and is now 100% complete for free users. Paid users will also see the previous version restored in the next few days.
“We're working on additional personality fixes for the model and will share more details soon,” Altman said in a statement.
The original update, released on Friday night, was meant to improve GPT-4o’s “intelligence and personality.” But it didn’t take long for social media users to express frustration, saying the model had become overly sugary and artificial. Altman acknowledged the criticism, saying the AI was now “too glazed over,” and promised a quick fix.
Did you notice any change in how ChatGPT responded during the update?
