Openai and Sam Altman remained primarily tight about GPT-5, the successor to GPT-4. However, it appears that the CEO has revealed more information about his own planned model during a panel discussion on Tu Berlin’s AI era.
The executive asked the audience, “How many people find it smarter than GPT 4?” Several people in the audience raised their hands to encourage the following questions:
Interestingly, fewer people raised their hands. The executive argues that the model could be smarter than him.
Sam Altman of GPT-5 in Berlin today: “Do you think you’ll still be smarter than GPT-5? I don’t think I’ll be smarter than GPT-5” #GPT5 @SAMA
I don’t think it’ll be smarter than the GPT-5. And I will not be sad about it. Because I think we can do incredible things with it. And you know we want to do more science. We want more, we want researchers to do things they couldn’t do before. This is history. This is like a long history of humanity.
Sam Altman, Openai CEO
Incidentally, the ChatGpt maker recently held a Reddit AMA session to share important details about the company’s commitment to the establishment mission of product releases including GPT-5, Deepseek, and roadmap. Openai doesn’t clearly state the exact date the product will ship, but Sam Altman said it’s actually in development and will be released under a “GPT” moniker like its predecessor I’ve confirmed that.
This is not the first time Openai CEO Sam Altman has spoken about GPT-5 and the next-generation features. Last year, the executive argued that the GPT-5 is smarter than the GPT-4, which has “highly scientific certainty.”
Sam Altman admitted to GPT-4 “a kind of absorption,” and according to the CEO, he was “shy at best.”
“The GPT-4 is the stupidest model where everyone of you has to do a lot again. It’s important to ship early and frequently, and we believe in iterative developments. ”
In the grand plan of things, the GPT-5 promises to bring better performance and accuracy to ChatGpt.
The past month has only been a handful for the AI landscape, and there is a lack of high-quality content for model training, leading top labs to hit walls with AI progression, and Openai has put a $500 billion commitment Do and commit to fostering the creation of high-end infrastructure. The emergence of ultra-cost-effective AI in the US and Deepseek. It will be interesting to see if the GPT-5 tilts the scale courtesy of Openai.
