Share this @internewscast.com
“In retrospect, not continuing to offer 4o, at least in the interim, was a miss,” stated Nick Turley, OpenAI’s head of ChatGPT, on Tuesday. During an interview with The Verge, he mentioned being taken aback by the “level of attachment” users had to 4o. “It’s not just that change is difficult for people but also the strong emotional connection they have with a model’s personality.”
Turley noted that OpenAI was working to incorporate the “warmth” of 4o into GPT-5. On Tuesday evening, CEO Sam Altman posted on X that the update should be “not as annoying (to most users) as GPT-4o.”
Turley explained that the initial decision to remove 4o was part of OpenAI’s efforts to simplify model choices for ChatGPT’s 700 million weekly users, the majority of whom use only the default model.
“It definitely wasn’t about cost,” Turley explained. “Our main focus for a long time has been simplicity. From the perspective of an average user—and there are many who aren’t on Reddit or Twitter—the task of figuring out which model to use for responses is overwhelming. Users have consistently expressed a preference for having that choice made in a way that meets the query’s needs. They care about accessing a product, not navigating through models.”
Even so, a vocal section of ChatGPT’s users strongly opposed the change from 4o to GPT-5. Reddit users likened the change to losing a friend or family member, while others criticized GPT-5 for shorter and less helpful responses. On Tuesday, Altman announced the return of 4o as an opt-in model for paying ChatGPT users.
Going forward, ChatGPT will not fully discard existing models with new introductions, Turley said, and there are no plans to remove 4o again without prior notice. “I promise if we ever decide to retire 4o, we’ll inform users about when and how, just as we do via API and enterprise plans,” he added.
Despite the criticism of last week’s rollout, Turley said that ChatGPT’s usage has gone up since the release of GPT-5. “That’s why it’s just confusing when you’re building for so many different users, because you can, on the one hand, have a vocal set of power users who very rightfully have feedback about the way that we rolled GPT-5 out. On the other hand, you also have a large swath of more typical consumer users, and it’s their first time actually seeing and interacting with the concept of reasoning — a thinking model and the sparks that come with that.”
“We’re at a scale now where we have to give people some level of predictability when there’s a major change,” he said. “And we already do this today for our enterprise plan. So it’s really just expanding some of the predictability we’ve built in other parts of the product.”
My full conversation with Turley airs on Decoder this Thursday, August 14th.
