OpenAI restores GPT-4o after user backlash over GPT-5 performance

New Delhi: OpenAI has responded to user complaints by bringing back GPT-4o following widespread dissatisfaction with its newer GPT-5 model, according to recent statements by CEO Sam Altman during a Reddit Ask Me Anything (AMA) session.

GPT-5, OpenAI’s latest large language model, was officially unveiled earlier this week and made available to over 700 million users worldwide. However, many free and paid ChatGPT users expressed frustration at the removal of GPT-4o, which they had relied on for their workflows. Several users described GPT-5 as “dumber” compared to the previous model and requested the option to revert.

In response to these concerns, Altman said, “We hear you all on 4o.” He confirmed that Plus users would be given the choice to continue using GPT-4o and that the company would monitor usage to decide how long to support it.

Criticism of GPT-5 has included complaints about the model’s colder “personality” and the abrupt removal of all legacy models in favour of a unified GPT-5 system. Some users reported being unaware that their preferred model would no longer be accessible.

A notable new feature of GPT-5 is its real-time router, designed to select the best model for a specific task. Altman acknowledged, however, that early issues with the router contributed to GPT-5’s perceived decline in performance. He stated that GPT-5 would “seem smarter starting today” and that OpenAI plans to “make it more transparent about which model is answering a given query.”

Currently, ChatGPT Pro, Team, and Enterprise users can switch back to GPT-4o via a “show legacy models” option in the app settings, but free users do not have access to the older model. When asked about the possibility of reinstating GPT-4.1, Altman said the company would consider it if there is sufficient demand from Plus users.

He concluded the AMA by saying OpenAI will “continue to work to get things stable and will keep listening to feedback.”