OpenAI CEO Sam Altman says the next AI model, GPT-6 will be smarter at remembering users and more personalized, making conversations feel more natural than before. The move comes just weeks after the launch of GPT-5, which drew criticism for feeling cold and limited in customization despite its technical strength.
Speaking to reporters in San Francisco, Altman said, “People want memory. People want product features that require us to be able to understand them.” But Altman did not share a release date for GPT-6, but he hinted it will arrive much sooner than the long wait between GPT-4 and GPT-5. He stressed that the new model won’t just answer questions. Rather it will learn to adapt to users and even let them build chatbots that reflect their own style and preferences.
Altman added that OpenAI has been collaborating with psychologists to guide the development of GPT-6, studying how people respond to long-term use and how it affects their overall well-being. While the findings haven’t been released yet, he suggested the company could share them in the future.
READ: OpenAI raises $6.6 billion, valuation hits $157 billion in latest funding round (
With GPT-6, OpenAI is aligning itself with the industry’s move from one-size-fits-all chatbots to more personalized AI agents. But the push also raises questions around privacy. Altman acknowledged that GPT-5’s temporary memory isn’t encrypted, leaving sensitive data at risk, and suggested that stronger protections, including encryption, “very well could be” added, though no timeline has been set. He emphasized that queries involving legal or medical information require robust privacy safeguards that aren’t fully in place today.
“It’s in society’s interest for people to get good medical advice … good legal advice,” he said. “And if you can get better versions of those from AI, you ought to be able to have the same protection for the same reason we decided you could get them from a doctor or a lawyer.” Looking further ahead, Altman also expressed excitement about brain-computer interfaces, calling them “a cool idea” and imagining a future where one could “think something and have ChatGPT respond.”
READ: OpenAI eyes $6 billion stock sale at $500 billion valuation (
Altman also stated that GPT-6 will follow guidelines, balancing neutrality with personalization, ensuring government use stays unbiased while still giving individuals the freedom to shape the chatbot to their own style and preferences. “I think our product should have a fairly center-of-the-road, middle stance, and then you should be able to push it pretty far,” Altman explained. “If you’re like, ‘I want you to be super woke’ — it should be super woke. If you want it to be conservative, it should reflect that as well,” Altman said.
Altman’s remarks come after a bumpy launch of GPT-5, which drew user complaints on social media for feeling colder, less engaging, and less useful than its predecessor. “I like the new one much better,” he said, acknowledging that the rollout had its flaws. He added that OpenAI had quietly implemented a tone update to GPT-5 that’s “much warmer.”