‘We totally screwed up’: Sam Altman explains what went wrong with GPT-5 launch

OpenAI CEO Sam Altman has accepted the mistakes made by the AI startup during its latest GPT-5 model rollout. Notably, soon after the latest model rollout, OpenAI had discontinued all its previous models leading to a barrage of crticism from its existing users and threats of subscription cancellations. 

The ChatGPT maker soon realized its mistake and started providing GPT-4o, the model previously running the chatbot, to ChatGPT Plus users while providing additonal rate limits for GPT-5 standard and Thinking models. 

OpenAI had claimed that its new model was leaps and bounds ahead of its previous offerings in areas like coding, reasing, accuracy, health, and multi-modal abilities. However, users complained that the new model gave shorter answers while showing less emotional depth in its responses than the previous offerings. 

In the immediate aftermath of the model launch, OpenAI had promised to make the new model ‘warmer’ while also bringing in a host of other changes. 

In a recent discussion with reporters on Thursday (as quoted by Bloomberg), the OpenAI chief executive admitted, “I think we totally screwed up some things on the rollout,”

“We've learned a lesson about what it means to upgrade a product for hundreds of millions of people in one day, and the differences in the kinds of attachment people have with this product versus previous products.” Altman added

Altman went on state that ChatGPT API traffic had doubled in the 48 hours corresponding the new model launch and the ChatGPT app usage was at a “complete high” in the days after GPT-5 release. He reportedly did agree that it was the wrong decision to deprecate all the older models. 

Last week OpenAI had also announced that it was touching 700 million weekly users which the company would have hoped would only rise with the launch of its new models. During its previous big GPT model launch earlier in the year, OpenAI had provided native image generation capabilities to GPT-4o which had led to a Studio Ghibli style craze worldwide and a major spike in ChaGPT users.

The same craze was nowheere to be seen this time around as social media soon filled up with users criticizing the approach taken by OpenAI with GPT-5 launch. The company had to even bring back the model picker after user backlash. Notably, removing the model picker completely was one of the big promises that Altman had made for GPT-5 in months preceding the launch.