Amid the recent backlash and mixed user feedback, OpenAI CEO Sam Altman has taken to social media to address the concerns following the launch of the new GPT-5 model. In a series of posts, Altman acknowledged that the company had “underestimated how much some of the things that people like in GPT-4o matter to them, even if GPT-5 performs better in most ways.”
Altman’s statement comes amid a growing list of complaints from users who feel the new model has a different personality, with many describing it as lacking the “warmth” and “emotional intelligence” of its predecessors. He said that users have “very different opinions on the relative strength of GPT-4o vs GPT-5” and recognised that a one-size-fits-all model isn’t effective in the world of AI bots.
OpenAI addressing GPT-5 complaints first
In the short term, OpenAI will prioritise finishing the GPT-5 rollout and making the model “warmer.” He also outlined a long-term vision for a more customisable user experience, stating, “We really need good ways for different users to customise things… I am confident we can offer way more customisation than we do now while still encouraging healthy use.” Adding a whiff of humour, he mentioned that some users love emojis, while others “never want to see one.”
Despite these plans, Altman warned of an upcoming operational challenges. “We are looking at a severe capacity challenge for next week,” he admitted, noting that the team is working to optimise systems and find more capacity. He promised to be transparent about the “tradeoffs” the company might have to make in the coming days, acknowledging that not everyone will be happy with the decisions.
“Thanks for your patience with us; we will continue to react and improve quickly!” he added.
Instant upcoming updates for GPT-5
OpenAI has announced several immediate changes to the GPT-5 rollout, particularly for its paid subscribers.
– The company plans to double the rate limits for ChatGPT Plus users, addressing a key complaint about restricted usage of the new model.
– OpenAI will allow Plus subscribers to continue using the GPT-4o model, though the duration of this option will depend on user demand.
– The company also addressed the technical glitches that have impacted user experience. Altman explained that a broken “autoswitcher” was responsible for making GPT-5 seem “way dumber” for a period, and he assured users that this issue is being fixed to ensure the correct model answers queries more often.
– In an effort to improve transparency and user control, OpenAI will make it clearer as to which model is answering a given prompt and will update the user interface to make it easier to manually activate the new “thinking” model.
