The Augmentation Question: What a Conversation Across Eight Time Zones Taught Me About AI's Real Opportunity
The best thing AI has done for me lately is introduce me to a human.
A few weeks ago, my post about the A24 and Google DeepMind partnership found its way to Tokyo, where Anna Skrypka, a Tokyo-based founder and writer who studies how Japan adopts AI, left the sharpest comment in a thread of hundreds: “That’s not hypocrisy. That’s what adoption actually looks like when trust is the variable.”
Trust is the variable. I couldn’t shake the line. So we started talking. Her evenings, my mornings, the Pacific between us. What emerged was one of the most clarifying conversations I’ve had about AI since this wave began.
Anna has spent 15+ years scaling creative technology companies across Japan and APAC, including nearly a decade at Autodesk and time inside Toyota’s Woven ecosystem. She’s watched technology adoption from inside the machine, across two business cultures that think about it very differently. And what she’s building now isn’t a company, it’s an intellectual movement she frames around a simple provocation: every civilization is being stress-tested by AI, and Japan is strangely built for it. Technology doesn’t decide what survives the transition, culture does. The question isn’t what AI can automate, but what we decide must stay human.
At the center of her thinking is a distinction I now can’t stop seeing everywhere.
Efficiency vs. augmentation
When a corporation meets AI, its first impulse is almost always efficiency: same output, fewer people, lower cost. It’s measurable, it’s immediate, and the market rewards it by Friday.
Augmentation asks a different question entirely: same people, radically extended capability. What could this team attempt now that was impossible last year?
Efficiency shrinks the denominator. Augmentation grows the numerator. One is subtraction dressed up as strategy. The other is the actual opportunity.
We’ve spent the past two months watching the efficiency-first crowd learn this the expensive way. Ford rehired 350 engineers. IBM tripling entry-level hiring. Klarna’s CEO admitting they “went too far.” Fifty-five percent of leaders who cut jobs for AI are now calling it a mistake. They optimized for efficiency and discovered, too late, that they had deleted capability.
Why the wrong prize keeps winning
Here’s the honest part: the efficiency impulse doesn’t win because executives are shortsighted. It wins because it’s legible. A CFO can model headcount reduction. Wall Street can price it. “Our people can now do things they couldn’t do before” doesn’t fit in an earnings deck, even when it’s the most valuable outcome by an order of magnitude.
That measurement gap, not bad intent, is why the wrong prize keeps winning. And it’s why the companies getting this right are the ones willing to measure something harder.
What augmentation actually looks like
It looks like the shot in Netflix’s The Eternaut that simply wouldn’t have existed, a building collapsing across Buenos Aires that the budget couldn’t touch until generative tools made it possible. Not fewer artists. A bigger canvas.
It looks like Dune: Part Two, where machine learning handled a thousand shots of glowing Fremen eyes so the artists could spend their hours on work that actually needed them. The team didn’t shrink. The ambition grew.
It looks like Gareth Edwards describing AI as “a second-unit director who is a billionaire on acid, it’ll do anything you ask.” Extension, not replacement. More capability per human, not fewer humans per task.
The view from Japan
This is where Anna’s vantage point matters most, because she’s watching this play out from a country that made the opposite bet from the start.
Japan faces a projected shortfall of 11 million workers by 2040. It cannot use AI for headcount reduction, there is no spare headcount to reduce. So Japanese adoption has defaulted to a different question: how do we preserve capability, and dignity, with fewer people? The government’s own framing is explicitly “human-centric AI”, augmenting human capabilities rather than replacing jobs. In elder care, the most acute crisis, the successful deployments are described in exactly these terms: augmentation over replacement, dignity preservation, technology facilitating human connection rather than substituting for it.
Anna’s observation, which has stuck with me since our first conversation, is that this isn’t just demographics, it’s all about values. Japanese business culture treats the dignity and worth of its people as something technology should serve, not spend. The demographic pressure didn’t create that instinct. It revealed it.
None of this means Japan is uniformly comfortable with AI. This is the country where Studio Ghibli’s co-founder famously called AI-generated animation “an insult to life itself.” The reverence for craft cuts both ways. But that’s precisely the point: a culture that takes craft seriously ends up asking better questions about what the technology is for.
The comparison is uncomfortable for the American approach. The U.S. reached for AI amid labor abundance and used it to cut. Japan reached for AI amid labor scarcity and used it to extend. Same technology, opposite question. And the rehiring wave suggests Japan’s question was the right one all along.
The augmented society
I’ve written before about who prospers in the AI era. In my view, it’s people who supply intention, standards, and judgment. Anna’s framing is the organizational answer to that same question. If humans are the source of intention, then the right corporate AI strategy is obvious: extend those humans. Make them more capable, not more optional.
That’s the augmented society she’s arguing for. Not a smaller workforce doing the same work cheaper. A workforce doing work that used to be impossible.
Eight time zones apart, over weeks of conversation, that’s where we landed. The question worth asking about your company’s AI strategy isn’t how much it saves. It’s this: is it making your people more capable or just making fewer of them?