The World's Smartest People Acknowledge AI Can Do What They Do

Something is shifting in the AI conversation. The world's most elite technologists, scientists, and business leaders are quietly admitting that AI is now doing what they do, at a level of competence they can no longer dismiss.

Former Dropbox CTO and early Facebook engineer Aditya Agarwal captured the mood after a weekend of coding with Anthropic’s Claude. He wrote, “We will never, ever write code by hand again," and that "something I was very good at is now free and abundant."

OpenAI CEO Sam Altman had a similar moment after building an app with OpenAI's Codex. The tool suggested features better than what he came up with himself. He said it made him feel "a little useless."

To understand what all of this signals, and why it matters to people other than investors, I talked it through with SmarterX and Marketing AI Institute founder and CEO Paul Roetzer on Episode 196 of The Artificial Intelligence Show. 

The Move 37 Moment for Knowledge Workers

The term "Move 37" traces back to the famous AlphaGo documentary, when DeepMind's AI made a move in the ancient game of Go that stunned even the world's top players. It was the moment humans realized a machine had surpassed them.

Roetzer has been tracking this same realization among knowledge workers.

"I define it as the moment when you realize AI is better than you at what you do," says Roetzer.

For Roetzer, this started in 2022 with early image generation tools and the arrival of ChatGPT. But what's different now is who's having the realization, and how loudly they're saying it.

An Emergency Meeting in Princeton

The most striking example came from David Kipping, an associate professor of astronomy at Columbia University and founding director of the Cool Worlds Laboratory. On his podcast, Kipping described attending what he called "one of the most impactful scientific meetings I've ever been to.” It was held at the Institute for Advanced Study in Princeton, NJ, which is one of the world’s most elite intellectual institutions. It’s the place where Einstein and Oppenheimer worked.

What stunned Kipping wasn't the presentation itself. It was that every scientist in the room was saying the same thing.

The lead faculty member told the group that AI models could already do roughly 90% of what he could do. The consensus in the room: AI had achieved what they called "complete coding supremacy" over humans, and was approaching comparable levels in analytical reasoning, problem-solving, and mathematics. This conversation happened among people in a room where, as Kipping put it, you'd be hard-pressed to find a higher average IQ anywhere on Earth.

Roetzer says the Princeton meeting is a signal we shouldn't ignore.

"The smartest people you can think of are worried about this,” Roetzer says.

AI Becomes a Business Negotiating Tool

These revelations are happening at the highest echelons of business, too.

Goldman Sachs announced it is deploying Anthropic's AI to automate trade accounting and client onboarding, after six months of embedded work with Anthropic engineers. Client onboarding times dropped 30%. Developer productivity jumped more than 20%. Over 12,000 Goldman developers and thousands of back-office staff are now using Claude daily.

And in what might be one of the first concrete examples of AI being used as leverage in fee negotiations, KPMG International pressured its auditor Grant Thornton to cut fees by 14%, arguing that AI efficiencies should reduce the cost of audit work. The message was blunt: if AI makes the work faster and cheaper, the price should reflect that.

AI Is Keeping Pace on the Hardest Problems

What makes this moment different from previous waves of AI hype is the caliber of people having these revelations. These aren't casual users impressed by a chatbot writing an email.

"These are people working on the hardest problems in humanity," says Roetzer. "And if they're seeing advancements like this, to where in an internal private meeting with no tech bros, no Silicon Valley VCs in the room, they're admitting to each other, this thing is on par with me, 160-plus IQs, that's hard to comprehend."

And yet, most others aren’t having this realization or even conversation.

"Society isn't even considering this more broadly,” Roetzer says. “And they won't know how to cope with it and how to move forward."

The people closest to this technology are processing so many emotions: wonder, sadness, disorientation, excitement. Agarwal called it "a weird time" and said the experience was "showing me what it is like to be human again." Altman described feeling "happy but sad."

Almost everyone is feeling these emotions, Roetzer says, and wondering what this means to them and especially the future of their job.

“It’s an ongoing conversation,” he says. “We're not trying to present some prescriptive answer here about what to do about it, but you're not alone.”

So whether you're an ad copywriter, a top salesperson, an HR professional, or a professor, the Move 37 moment is coming for your field, if it hasn’t already. It helps to move through it together, Roetzer says.

“Let's go through the different emotions and let's figure out the path forward," he says.