So, this is kind of wild. Sam Altman, the CEO of OpenAI — the company behind ChatGPT — recently spilled the beans on how powerful their newest AI, GPT-5, actually is. And honestly, it left him feeling a bit lost. Like, he even compared creating this AI to the Manhattan Project — yeah, the same secret World War II program that made the atomic bomb. That’s some heavy stuff.

During a recent podcast, Altman shared that when testing GPT-5, the AI solved problems that even he couldn’t figure out. He said he felt "useless," which is pretty telling coming from the guy who helped build it. It’s not so much fear, but a strange kind of awe and maybe a bit of worry about what this means for all of us.

Why This Feels Different: The Manhattan Project Comparison

When Altman talks about the Manhattan Project, he’s not just being dramatic. That project changed humanity forever — it showed how science can unleash power that’s hard to control. Similarly, GPT-5 isn’t a bomb, but the kind of tech we’re building could change our world in ways we can’t fully predict.

He said moments like this happen sometimes in science, where you stop and ask yourself, “What have we done?” ChatGPT 5’s ability to answer tough questions, write code, and basically act like a super-smart expert has taken a huge leap from what we had before. Altman feels this AI might be crossing a line where its thinking starts to outpace humans in real and meaningful ways.

What’s So Special About GPT-5?

OpenAI says GPT-5 is a big step closer to something called AGI — Artificial General Intelligence — which means an AI that can think and understand like a human across pretty much any task. It can:

  • Write complex computer code quickly

  • Help with creative writing and problem-solving

  • Give expert-level advice like a PhD in your pocket

Altman even joked that going back to older models feels “miserable” because GPT-5 is just on a whole new level. Actually, it’s much faster and less likely to mess up facts or hallucinate nonsense compared to earlier versions.

But despite all this power, Altman says GPT-5 still isn’t perfect — it’s missing some key things like learning from the world in real time. So the journey to a truly human-like AI isn’t over yet.

Also read on Procapitas

Why Should We Care? The Bigger Risks and Questions

Here’s the thing — this isn’t just tech hype. As AI grows stronger, it raises real questions:

  • Are we prepared for machines that outthink us?

  • What rules should be in place before AI gets out of control?

  • How will these changes impact jobs, privacy, and society?

  • Could it be misused or cause harm before we understand the consequences?

Altman says there are “no adults in the room” when it comes to regulating this stuff — meaning the world’s leaders and companies might not be ready to handle AI this powerful responsibly.

What’s Next for OpenAI and the AI World?

OpenAI’s about to roll GPT-5 out to millions of users, blending it into all sorts of apps and services. Investors are watching closely — the company’s valuation might hit $500 billion soon.

But with so much power comes pressure to be careful, and Altman’s candid words give a rare peek into the anxiety behind this tech race. They’re trying to balance pushing AI forward with making sure it doesn’t spiral out of control.

For everyday folks, it means we’re entering a new chapter where AI feels smarter than ever and where the big question remains: Are we ready for what comes next?