OpenAI CEO Sam Altman has addressed the rocky launch of GPT-5, which was met with technical hiccups, user complaints, and even a presentation blunder.
Following its release, some users experienced slow responses and outages, while others were frustrated by the limited “mini” version offered to free and Plus accounts. Full GPT-5 access was initially restricted to Pro subscribers.
The launch event also sparked criticism over a misleading performance chart. A graph showing GPT-5’s error rates was visually inaccurate, prompting Altman to call it a “mega chart screw-up.” He explained that the mistake was the result of late-night work and human error. The chart was later corrected, and an apology was issued.
Another flashpoint was the removal of GPT-4o, a popular model valued for its creativity and conversational tone. Its sudden disappearance led to strong backlash from users. In response, Altman confirmed that GPT-4o will return as an option for Plus subscribers while OpenAI evaluates how long to keep it available.
The company says it will continue refining GPT-5 and listening to user feedback as it balances innovation with reliability.
If this article helped you, please consider supporting our work. Every small contribution keeps Abijita.com independent and running.
