During a Reddit ask-me-anything<\/a> session on Friday, OpenAI CEO Sam Altman and key members of the GPT-5 team were peppered with questions about the new model and requests to bring back its previous model, GPT-4o.<\/p>\n They also asked Altman about the most embarrassing \u2014 and perhaps funniest \u2014 snafu in the presentation, the \u201cchart crime.\u201d<\/p>\n One of the new features that GPT-5 rolled out is a real-time router<\/a> that decides which model to use for a particular prompt, either responding quickly or taking additional time to \u201cthink\u201d through answers.<\/p>\n But multiple people in the AMA on the r\/ChatGPT Reddit complained<\/a> GPT-5 wasn\u2019t working as well for them as 4o did.\u00a0Altman said the reason GPT-5 seemed \u201cdumber\u201d was the router wasn\u2019t working properly when it was rolled out Thursday.<\/p>\n \u201cGPT-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. Also, we are making some interventions to how the decision boundary works that should help you get the right model more often. We will make it more transparent about which model is answering a given query,\u201d Altman promised.<\/p>\n Still, people on the AMA lobbied so hard to bring 4o back for Plus subscribers that Altman promised to at least look into that. \u201cWe are looking into letting Plus users to continue to use 4o. We are trying to gather more data on the tradeoffs,\u201d he wrote. <\/p>\n And Altman also promised, \u201cWe are going to double rate limits for Plus users as we finish rollout.\u201d\u00a0This should give people a chance to play and learn the new model, adopt it to their use cases without worry of running out of monthly prompts.<\/p>\n Techcrunch event<\/p>\n \n\t\t\t\t\t\t\t\t\tSan Francisco<\/span> Predictably, he was also asked about the wildly inaccurate chart the team presented during the live presentation that quickly became the butt of many \u201cchart crime\u201d jokes<\/a>. The chart presented a lower benchmark score with a much taller bar.\u00a0<\/p>\n Altman didn\u2019t answer questions about the chart during the AMA, but on Thursday he did call the chart a \u201cmega chart screwup\u201d on X<\/a>. Others noted the charts in the published blog post were correct<\/a>.<\/p>\n But the damage was done. Jokes ensued about using GPT for charts in a corporate presentation. GPT-5 reviewer Simon Willison, who had early access and generally liked the model\u2019s performance, also pointed out <\/a>that turning data into a table was \u201cgood example of a GPT-5 failure.\u201d<\/p>\n In any case, Altman promised fixes to the items that seemed to concern people the most. He ended the AMA with a promise: \u201cWe will continue to work to get things stable and will keep listening to feedback.\u201d<\/p>\n<\/div>\n
\n\t\t\t\t\t\t\t\t\t\t\t\t\t|<\/span>
\n\t\t\t\t\t\t\t\t\t\t\t\t\tOctober 27-29, 2025<\/span>\n\t\t\t\t\t\t\t<\/p>\n<\/p><\/div>\n<\/p><\/div>\n<\/div>\n![\"OpenAI's](\"https:\/\/techcrunch.com\/wp-content\/uploads\/2025\/08\/GPT-5-chart-crime-chart.png?w=680\" "\\"\\"")