Stop asking what AI will do to your job — start asking what constraint your job actually exists to solve

Sangeet Paul Choudary: There are really two reasons I felt I had to write this book now.

The first is that there’s so much noise about AI — but almost all of it is about playing the same old game faster. Most discussions focus on automating tasks, improving efficiency, or boosting productivity. What almost no one is asking is: how does AI actually change the game itself?

How will it reshape the kinds of companies that emerge, how products are imagined, and how power shifts across value chains? That larger, systems-level conversation — about how AI restructures the economy itself — was missing, and I wanted to bring it to the foreground.

The second reason is that our current discussions about AI are very fragmented. On one side, we talk about what AI will do to my job. On the other, we talk about what it will do to my industry. I wanted to show that these are the same story. AI doesn’t replace jobs just by automating tasks — it changes the industrial and organizational systems that give those jobs meaning in the first place. Once those systems shift, every role within them is redefined.

And this ties into the broader arc of my work. I’ve always been interested in how technological shifts move the goalposts — how they change where advantage lies. Companies don’t lose because they fail to execute; they lose because they keep executing brilliantly in a game whose rules have already changed.

So this book is about helping people see that shift clearly — not just how AI helps you work faster, but how it forces all of us to play differently.

Extract from Reshuffle, Chapter 1:

We live in an era shaped by what everyone calls exponential technologies. The challenge, pundits say, is that most of us struggle to grasp exponential effects because we tend to think in straight lines. We expect change to be gradual and predictable.

To illustrate this, a parable commonly used is that of a farmer who approaches a king, asking for a single grain of rice, doubled on each square of a chessboard – just 64 squares. The king, amused, agrees. The request seems modest – one grain for the first square, two for the next, then four, then eight. Even after 20 squares, it’s barely a sack. But soon, the numbers start to balloon: by the 20th square, the king owes over a million grains; a trillion by the 40th. By the end of the board, the total exceeds all the rice in the kingdom.

This story is often used to illustrate the power of exponential growth: slow and deceptive at first, then overwhelming. The magic is attributed to compounding. Small gains stack up fast; then suddenly, you’re buried in riches.

Compounding is an easy story to tell. It’s why the most common parable of exponential growth involves nothing more than a simple doubling pattern: a story of grains of rice on a chessboard and a king who didn’t do the math.

And yet, when we look at how economic systems actually evolve, compounding turns out to be a surprisingly incomplete explanation. Compounding might explain how technologies evolve – faster chips, improving models, and more data. However, economic and social systems don’t scale simply because things become bigger or faster. They scale because things start working together in fundamentally new ways. This is the second untold story of global trade.

At first, the impact of containerization seemed obvious. Ports got more efficient, ships spent less time waiting, and dock work declined. But these were only first-order effects – the initial incremental gains you’d expect from a compounding model.

The real transformation came later, not through scale, but through scope. Standardized containers and standardized contracts enabled intermodal transport. With that, freight became faster, cheaper, and more reliable. And that reliability broke the logic of vertical integration. Firms could specialize and outsource. And as companies specialized, components improved. Improving components led to greater product innovation. And the cascade continued. High-knowledge work was separated from high-efficiency production, inventory management changed, and trade accelerated. This didn’t happen because of compounding. It happened because of cascading effects – one solved coordination problem unlocking the next layer of opportunity.

This is the secret to exponential system change: not compounding acceleration, but cascading coordination.

Compounding is about scale; cascading effects drive scope expansion. The first breakthrough may look like cost savings, but the second creates new industries. By the time the third breakthrough materializes, all economic activity is restructured. With each solved coordination problem, new layers of activity are unlocked and the system’s scope expands.

This is the untold story of global trade, and it’s the real story of AI. AI might follow a compounding trajectory in terms of raw capability, but its true impact will come from how well systems coordinate to harness it. The idea that AI will suddenly leap into superintelligence distracts us from the real work of evolving our systems to keep up with today’s capabilities. What matters isn’t the scale of processing power. What really matters are the cascading effects that emerge when coordination unlocks entirely new possibilities.

SPC: The most important trend that’s only just beginning to unfold is the commoditization of knowledge work.

AI is starting to do to the knowledge economy what industrial automation once did to manufacturing — it’s making previously scarce expertise cheap, fast, and interchangeable. Whether it’s the research cycles of drug development, geo-exploration in mining, or even everyday consulting and legal work, the underlying pattern is the same: what used to be a craft is becoming a computation.

As that happens, value doesn’t disappear — it moves. It migrates to new parts of the value chain: to whoever can coordinate, curate, and integrate all these now-commoditized capabilities into coherent systems and guaranteed outcomes. That shift is just beginning, and most people still don’t see it.

I think that’s where the real opportunity lies — not in chasing AI to do what we already do faster, but in reimagining where advantage sits once AI makes expertise abundant.

SPC: Stop asking what AI will do to your job — start asking what constraint your job actually exists to solve.

Most of the popular advice out there — like “AI won’t take your job, but someone using AI will” — sounds empowering but misses the point. It assumes that simply adopting a new tool will protect you. It won’t.

Our jobs aren’t defined by the tasks we perform; they’re defined by the constraints we help overcome in a larger workflow or organization. When those constraints change, the nature and value of the job change with them.

Take the typist, for example. Typing wasn’t the real job — managing the cost of editing was. Before word processors, editing was slow and expensive, so accurate typing was valuable. Once the cost of editing collapsed, the constraint disappeared, and so did the typist’s role.

So, if you really want to future-proof yourself, don’t just reskill randomly or chase the next hot tool. Ask: what constraint does my work solve today, and how will AI change that constraint? The people who can see and adapt to those shifting constraints are the ones who’ll create the next generation of meaningful work.

SPC: The next big theme I’m exploring is the idea of structural uncertainty — and how to think in a world where the rules themselves keep changing.

For most of the past century, businesses have dealt with operational uncertainty: fluctuations in demand, supply shocks, or executional hiccups — problems you could manage with buffers, planning, or agility. But today, we’re in a world where the structure of industries is shifting under our feet. The game itself is being rewritten — who the players are, what counts as value, and how advantage is built.

That’s structural uncertainty — when the very frame you’re operating within is unstable. And most organizations still try to manage it with operational tools like Agile or Lean, which were designed for a very different world. My next book is about this shift: how to recognize structural uncertainty, why our existing playbooks fail, and what a new toolkit for navigating it might look like.

Thank you Sangeet Paul Choudary

Thank you Bertrand Jouvenot

The book: Reshuffle, Sangeet Paul Choudary, 2025