GPT-5 Backlash Shows Some Users Are Getting Hooked on Specific Models

OpenAI CEO Sam Altman’s big GPT-5 announcement was drowned out by users clamoring to keep their access to older models.

Less than 24 hours after OpenAI announced it would discontinue 4o, Altman and his team signaled a retreat. The fierce responses show that OpenAI’s userbase has developed strong attachments to its various AI models, an achievement in and of itself, but one that poses a challenge to the company as it looks to convince people to shift to newer models.

“If you have been following the GPT-5 rollout, one thing you might be noticing is how much of an attachment some people have to specific AI models,” Altman wrote on X on OpenAI Sunday. “It feels different and stronger than the kinds of attachment people have had to previous kinds of technology (and so suddenly deprecating old models that users depended on in their workflows was a mistake).”

Users have molded their ChatGPTs into quasi-therapists, life coaches, and friends. Altman previously said that one user said their GPT gave them more encouragement than a parent ever had. Other tech companies have annoyed and angered their users by changing hardware or interface. But Altman’s rival CEOs never dealt with emotional bonds that mimic real life.

The personality changes between models were particularly jarring for some.

“GPT5 colder personality doesn’t help either, it’s just sad,” one user wrote during the Reddit AMA.

Another added after Altman said the older model was coming back for paid users, “Thanks, Sam—bringing back 4o is a fantastic first step. Please consider keeping it around long-term. Many of us rely on it every day, not just for accuracy or logic, but for the warmth newer models sometimes miss.”

Nick Turley, OpenAI’s head of ChatGPT, said that the Reddit comments showed “just how polarized users could be.”

“It’s not just change that is difficult for folks, it’s also the fact that people can have such a strong feeling about the personality of a model,” Turley told The Verge.

Nicole DiNicola, VP of Marketing at Smartcat, said one of the mistakes OpenAI made was abruptly removing access to earlier models without warning.

Related stories

“The expectations have gotten so much higher for anything AI-related that it is much easier to be disappointed,” DiNicola told Business Insider.

Turley said the “level of passion” some users had for 4o led him to rethink some things. He also said that OpenAI announced before GPT-5 that users could select from four personalities for their chatbots: cynic, robot, listener, and nerd.

“When I see people saying, ‘Hey, this is my only and best friend,’ that doesn’t feel like the type of thing that I wanted to build into ChatGPT.” Turley said. “That feels like a side effect, and it’s therefore worth taking seriously and studying closely, and that’s what we’re doing.”

Harvey Hu, cofounder of General Agency AI, an applied research lab, said he found some of GPT-5’s quick responses to be “slightly more stupid than 4o.”

It’s not just functionality that can make users attached.

“There may be a certain emotion, certain ways of prompting that you train yourself to learn, and if the certain model you got used gets sunsetted that could be definitely emotional thing for the users,” Hu told Business Insider.