The company said the change will take place alongside the previously announced retirement of GPT-5 Instant and GPT-5 Thinking models. The move applies only to ChatGPT, with no changes planned for the company’s application programming interface (API) at this stage.
GPT-4o, which had been temporarily removed and later restored during the rollout of GPT-5, was given special consideration due to user feedback. OpenAI said some Plus and Pro subscribers had asked for more time to transition away from the model, citing its conversational tone and usefulness for creative ideation.
See also: OpenAI Rolls Out AI Education Framework With Early Group of Nations
“We brought GPT-4o back after hearing clear feedback from a subset of Plus and Pro users, who told us they needed more time to transition key use cases, like creative ideation, and that they preferred GPT-4o’s conversational style and warmth,” the company said in its press release.
OpenAI said those insights were incorporated into newer versions, including GPT-5.1 and GPT-5.2, which feature expanded customization options such as adjustable tone, warmth and enthusiasm. The company added that the vast majority of ChatGPT usage has now shifted to GPT-5.2, with only about 0.1% of users still selecting GPT-4o on a daily basis.
See also: OpenAI Introduces Prism, AI-based Workspace for Scientific Writing and Collaboration
The company said it is continuing to refine ChatGPT’s personality and creativity, while also reducing what users have described as unnecessary refusals or overly cautious responses. It is also developing a version of ChatGPT designed for users aged 18 and over, supported by age-prediction systems for younger users in most markets.
“Retiring models is never easy, but it allows us to focus on improving the models most people use today,” OpenAI said, acknowledging that some users may find the change frustrating.
Source: OpenAI