In a candid, unprecedented interview on comedian Theo Von’s hit podcast This Past Weekend, OpenAI CEO Sam Altman spoke candidly about a moment that shifted his perception of his own intelligence—and humanity’s future.
The tech entrepreneur shared a dramatic experience with OpenAI’s as-yet-unreleased generative model, GPT-5. “I had a problem that had been puzzling me,” Altman admitted. “I didn’t know what to do with it. I plugged it into GPT-5, and it just. answered it. Perfectly. Not great, but perfectly.”
The experience, he explained, was both incredible and unnerving. “It was an odd sensation. I felt kind of redundant, in relation to the AI. It was a genuine moment of humility.”
AI That Surpasses Its Creator
Altman’s comments verge on what may be the biggest jump so far in the generative AI race. While GPT-5 has not yet been publicly released, Altman indicated that it is the culmination of multiple major developments OpenAI has been working on over the last few years.
“It combines a lot of our technology,” he explained, proposing a convergence of multimodal capabilities (text, image, audio, and potentially video), richer reasoning structures, and larger context windows—all of which AI researchers have been consistently moving toward. “It’s not simply faster or more correct—it thinks differently,” Altman added.
Though he was tight-lipped about a launch date, the eagerness of Altman was evident. His statement has renewed speculation within tech communities that GPT-5 could be released ahead of schedule, possibly in stages or for selected developers before a wider rollout.
Between Anxiety and Wonder
What made this exchange so remarkable wasn’t the technical jargon—it was Altman’s vulnerability. For one of Silicon Valley’s most powerful players to acknowledge feeling “useless” in light of his own invention is a new note in the AI conversation.
“I believe there’s this moment,” Altman said, “where you see that AI is not just copying human intelligence. It’s beginning to overtake it in certain areas. And that’s really a strange thing to sit with.”
Von, who was famous for his quirky but profoundly human interviews, pushed Altman on the emotional resonance of such advancements. Did he fear what he was creating?
“Yeah. Sometimes,” Altman acknowledged. “But I also believe that fear is important. It’s a reminder that we are venturing out into the unknown. And it makes us be cautious.”
This combination of enthusiasm and existential prudence is one Altman has discussed in the past, but never so personally, so candidly. He stressed that OpenAI remains committed to alignment research—making sure that more powerful AI systems remain safe, controllable, and good for society.
Rethinking What It Means to Know
Altman’s anecdote also raises an age-old philosophical question: What happens when machines are able to know things that people can’t?
“GPT-5 did not just provide me an answer. It provided me the answer. With clarity, structure, and reasoning that I could not immediately do myself,” Altman explained. “And that got me thinking—what’s the role of human intelligence in a world where AI is able to do this?”
It’s a question that has terrorized researchers for decades. Conventionally, we’ve quantified knowledge by what humans are able to understand. But if computers can compute solutions or provide insights that we can’t—even after going over their results—does that indicate we’ve created something genuinely extraterrestrial?
“There will be a moment,” Altman said, “when we’ll have to rely on the AI being correct, even when we don’t know exactly how it arrived there. And that’s both exhilarating and very uncomfortable.”
Fame, Friction, and the Future
Altman also spoke candidly about the pressures of leading a company at the epicenter of one of the most transformative technologies of our time. Since the launch of ChatGPT in 2022, Altman has become a public figure far beyond the tech world. But with that fame has come scrutiny—over OpenAI’s partnerships, its direction, and the future it is shaping.
“It’s strange. I didn’t set out to be a public person,” he said. “But I do think transparency matters. I want people to see who’s behind the decisions, especially when the stakes are this high.”
He also touched on some of the recent tension inside OpenAI—leadership shifts, conflicting visions, and pressure from investors and regulators. “We’re a team of passionate people who care deeply. Sometimes we disagree. But I think disagreement, handled well, makes the product and the company stronger.”
Altman’s discussion with Von was not a pitch or PR stunt—it was an insight into the mind of someone who is wrestling with the implications of creating something that could redefine what it means to be smart, creative, or even human.
“AI is not going to replace us,” he stated towards the end of the podcast. “But it is going to make us evolve—to ask more thoughtful questions, to be more reflective, to rethink what we value in human experience.”
Whether that development brings us to utopia or something more complex is uncertain. But one thing is certain: even the individual guiding the AI revolution is awed—and sometimes terrified—by what lies ahead.
“Sometimes,” Altman said, “I think we’re not just building tools. We’re building mirrors. And they reflect back to us who we are—and who we may become.”
