Sam Altman Responds to GPT-5 Launch Issues and Promises Improvements
OpenAI’s Sam Altman Responds to GPT-5 Rollout Challenges
OpenAI CEO Sam Altman and members of the GPT-5 team recently participated in a Reddit AMA, addressing community concerns over the launch of GPT-5 and the temporary removal of GPT-4o. Users voiced frustrations about performance and transparency, prompting Altman to acknowledge the "bumpy" rollout and promise rapid improvements.
Why Did GPT-5 Seem “Dumber”?
Many Reddit users reported that GPT-5 wasn’t performing as well as expected, especially compared to the previous GPT-4o model. Altman explained that a technical issue with the new real-time router—meant to dynamically select the best model for each prompt—led to subpar responses during the initial rollout. He stated:
- "GPT-5 will seem smarter starting today. Yesterday, we had a sev and the autoswitcher was out of commission for a chunk of the day, and the result was GPT-5 seemed way dumber."
- OpenAI is now making the model selection process more transparent, so users will know which version is responding to their queries.
Will GPT-4o Return?
Given the strong support for GPT-4o, especially among Plus subscribers, Altman confirmed that OpenAI is exploring options to let users continue accessing the previous model. He said, "We are looking into letting Plus users continue to use 4o. We are trying to gather more data on the tradeoffs." This decision comes as the team works to balance feature improvements and user preference.
Better Access for Plus Users
To support users adapting to the new model, Altman also promised to double the rate limits for Plus subscribers as the rollout completes. This will give subscribers more flexibility to test, learn, and integrate GPT-5 into their workflows without worrying about hitting monthly usage caps.
The “Chart Crime” Incident
During GPT-5's live launch, a chart shown on screen sparked immediate backlash for misrepresenting benchmark data—a mishap quickly dubbed a "chart crime" on social media. The chart depicted a much taller bar for a lower score, leading to jokes and criticism throughout the tech community.
Altman addressed the incident on X, calling it a "mega chart screwup." He clarified that accurate charts were published in OpenAI’s official blog post, but the initial presentation had already circulated widely online.
Community Feedback Drives Change
Throughout the AMA, Altman and the OpenAI team emphasized their commitment to listening to user feedback and quickly addressing technical issues. Altman concluded the session by stating, "We will continue to work to get things stable and will keep listening to feedback."
Looking Ahead
As OpenAI continues to refine GPT-5 and potentially brings back GPT-4o for select users, the company is focused on improving reliability, transparency, and user experience for its growing community.
