Altman acknowledged complaints that GPT-5 felt “colder” and less helpful than its predecessor, noting the company has already updated the model’s tone.
OpenAI is already looking past its latest model. Just weeks after launching GPT-5, Chief Executive Sam Altman said development of GPT-6 is underway and will advance on a faster timeline than earlier releases.
According to a CNBC report, Altman acknowledged complaints that GPT-5 felt “colder” and less helpful than its predecessor, noting the company has already updated the model’s tone, while talking to reporters in San Francisco. “I like the new one much better,” he said, noting OpenAI has quietly updated the tone of GPT-5 to make it “warmer.” At the time of writing, retail sentiment around OpenAI on Stocktwits trended in ‘neutral territory.
The company’s next model, Altman said, won’t simply respond to prompts but will adapt to individual users. He pointed to memory as the critical feature for making ChatGPT “personal,” allowing the system to retain preferences, routines, and quirks over time. OpenAI has been working with psychologists to help shape the product, measuring how users feel and tracking well-being, though Altman said that research hasn’t been made public.
GPT-6 will also be built to comply with a Trump administration executive order requiring AI systems used by federal agencies to be ideologically neutral and customizable. Altman said the company intends to offer a default “middle stance” while letting users shift the model toward conservative or progressive tones. “If you’re like, ‘I want you to be super woke’ — it should be super woke,” Altman said. “If you want the opposite, it should do that too.”
The accelerated push comes as competitive pressures from rivals, including Amazon-backed Anthropic, Google DeepMind, and Elon Musk’s xAI, are ramping up.
Read also: Trump Says ‘It’s Possible Putin Doesn’t Want To Make Deal’ – Urges Zelenskyy-Putin Meeting First
For updates and corrections, email newsroom[at]stocktwits[dot]com.<