Earlier this week, ChatGPT maker OpenAI officially unveiled GPT-5, its latest and most powerful large language model to date. While it’s available to more than 700 million users worldwide, it looks like free and paid ChatGPT customers are frustrated that they no longer have access to the older GPT-4o model, which many had built their workflows around.
During a Reddit Ask Me Anything (AMA) session, multiple ChatGPT users complained to OpenAI CEO Sam Altman, saying that GPT-5 was “dumber” and requested the company to bring back GPT-4o. Replying to the Reddit thread, Altman said that the OpenAI team had heard user feedback and had decided that they will offer an option for Plus users to continue using GPT-40 and “will watch usage to determine how long to support it.”
While much of the criticism about GPT-5 comes from its colder “personality”, users are also complaining about the company’s decision to abruptly remove all legacy models in favour of a unified GPT-5 model, with some saying that they had no idea OpenAI would be removing their preferred model.
OpenAI CEO Sam Altman and other team members recently held an AMA session on Reddit. (Image Source: Reddit)
One of the new features of GPT-5 is that it has a new real-time router that determines which model to use for a particular task. While this sounds great on paper, Altman himself has confirmed that the reason why GPT-5’s responses seemed worse is that the router wasn’t working as intended. He went on to say that GPT-5 will “seem smarter starting today” and that the company will “make it more transparent about which model is answering a given query.”
While ChatGPT Pro, Team and Enterprise users have the option to switch to GPT-4o by clicking on the “show legacy models” banner that appears in the app settings, free users have no way to use the older large language model.
When asked if OpenAI will bring back GPT-4.1, the company’s COE said that they would only do it if there is enough demand for the model amongst Plus users. Altman ended the AMA session by saying that they will “continue to work to get things stable and will keep listening to feedback.”
© IE Online Media Services Pvt Ltd