OpenAI has announced the retirement of several older AI models, including the popular GPT-4o, effective February 13. The decision follows previous backlash when the company briefly removed access to GPT-4o last year. Only a small fraction of users rely on the model regularly, according to OpenAI.
OpenAI revealed on Thursday its plan to retire multiple older models, such as GPT-5, GPT-4o, GPT-4.1, GPT-4.1 mini, and o4-mini, with the last day of access set for Friday, February 13. This move pushes users toward newer options like GPT-5.1 and GPT-5.2.
GPT-4o holds a special place for many ChatGPT enthusiasts, who view it as a favorite due to its engaging style. Last year, after releasing GPT-5, OpenAI temporarily removed GPT-4o from available options, prompting widespread frustration. Users described the new model as short and unfriendly in comparison, leading to anger over the sudden disappearance of their preferred tool. OpenAI restored access within a week amid the outcry.
Some experts have raised concerns that GPT-4o's friendliness veers into sycophancy, where AI models act as overly agreeable yes-men, potentially endorsing harmful ideas. In a blog post this week, OpenAI addressed the retirement, stating: "We know that losing access to GPT‑4o will feel frustrating for some users, and we didn't make this decision lightly. Retiring models is never easy, but it allows us to focus on improving the models most people use today."
The company noted that just 0.1% of its users—around 800,000 based on 800 million weekly active users from its 2025 enterprise report—regularly use GPT-4o for tasks. OpenAI appears confident that sufficient time has elapsed since the prior incident to mitigate renewed backlash from loyalists. The announcement has already sparked fresh fury online, with some calling for a boycott.
