Why The Future Is Always Overpriced may be a book for this exact moment, Forget the lazy question: “Is AI a bubble?”
It is almost certainly the wrong question, the more dangerous possibility is far more uncomfortable: AI could genuinely transform the world, and investors could still massively overpay for that transformation.
Both things can be true, in fact, history suggests they often are.
That is the central provocation behind The Future Is Always Overpriced: The Capital Spiral Metaphor from Railway Mania to the Financialisation of Intelligence, a book about technology, capital, speculation and the strange way modern economies repeatedly attempt to purchase the future before anyone knows what that future is worth.
And right now, this is no longer an abstract argument, the future is being ordered, the concrete is being poured, the chips are being manufactured, the data centres are rising,. the electricity is being demanded. the capital is already moving.
The only unanswered question is whether the economic future arrives quickly enough to justify the price being paid for it.
The Biggest Bubbles Don’t Necessarily Begin With Lies
We like bubbles to be stupid; it makes them easy to understand.
Someone invents nonsense. Greedy financiers sell it. Gullible investors buy it. Prices explode. Reality arrives. Everyone discovers the emperor was naked, but history’s most important investment booms are often much more dangerous than that.
They begin with something real. Railways really did transform transportation, electricity really did reorganise industrial production, automobiles really did reshape cities, the internet really did transform communication, commerce and information, and artificial intelligence really may transform the economics of cognition.
That is precisely the problem.
As the book argues, transformative technologies possess an extraordinary ability to attract excess capital because the underlying technological story may actually be true.
The progression is seductive: The technology works. Then: The technology will become important.
Then: The technology will become enormous, and finally comes the leap markets repeatedly make:
Therefore, almost any amount of capital invested today must eventually earn an attractive return.
It does not follow, being right about the future is not the same thing as being right about the price of the future. That distinction sits at the heart of the book.
History is filled with people who correctly predicted what the world would become — and still lost fortunes trying to own it.
Welcome to the Capital Spiral Metaphor
The book introduces a framework it calls the Capital Spiral Metaphor: a genuine technological opportunity appears, capital notices, investment arrives, infrastructure gets built, costs fall, adoption increases, success creates evidence, evidence attracts more capital, competition intensifies, expectations rise, financial markets invent new ways of financing the opportunity, more capital arrives, more infrastructure gets built, eventually, the system can cross an invisible line.
Capital is no longer merely financing the future, capital is financing the expectation of the future, and those expectations begin affecting the economy that is supposed to validate them.
This is why the book calls the phenomenon a spiral rather than a cycle, a cycle returns to where it started, a spiral does not.
The psychology repeats greed, fear, extrapolation, FOMO, competitive pressure, but underneath it, infrastructure accumulates, technology advances and finance learns new tricks.
The psychology is cyclical. The economy is cumulative.
The Railway Investors Were Right. Many Still Lost.
Go back to Britain in the 1840s, railways were not a fantasy, they worked spectacularly. They collapsed journey times, moved freight, connected towns and changed the economic geography of Britain. Capital flooded towards them, railway projects multiplied, tracks were laid, fortunes were imagined, then came the reckoning, but something extraordinary happened after the financial damage, the railway tracks did not disappear.
Share certificates could collapse, companies could fail, investors could lose fortunes, but the infrastructure survived, and society inherited it. That pattern appears repeatedly throughout the book, electricity, automobiles, the infrastructure excesses of the twentieth century, the dot-com boom, housing and securitisation, Crypto and now artificial intelligence.
The book’s historical journey is deliberately cumulative: from selling claims on tomorrow, to financing enormous physical networks, to financing consumer demand, to turning cash flows into collateral, to making financial claims programmable — and ultimately to asking whether intelligence itself is becoming financeable.
That is where the argument becomes much bigger than another history of financial bubbles.
A Crash Can Destroy Capital Without Destroying the Future
Here is one of the book’s most unsettling ideas:
A disastrous investment can create an extraordinarily useful asset.
Imagine infrastructure that costs $10 billion to construct but cannot generate enough revenue to justify that valuation. The company fails. Creditors take losses. The asset changes hands for $4 billion. Nothing physical has changed. The machines still work. The network still operates. The infrastructure still exists, but suddenly the economics are completely different because the next owner did not pay the original price. The financial claims have been destroyed, the productive capacity has survived.
That distinction changes how we should think about technological booms.
The railway can survive the railway investor. The fibre network can survive the telecommunications company. The data centre may survive the company that financed it. The infrastructure does not remember the share price, and once yesterday’s investors have paid the enormous fixed cost of constructing tomorrow’s infrastructure, the next generation of entrepreneurs may inherit it far more cheaply.
That leads to one of the Capital Spiral’s deepest paradoxes:
Scarcity creates the boom.
The boom creates abundance.
Abundance destroys the scarcity that justified the boom.
Yet that same abundance may be exactly what allows the technology to transform society.
Investors can lose, consumers can win, the revolution can continue.
Then Comes AI
This is where history suddenly stops feeling historical.
Because artificial intelligence appears to combine several previous technological revolutions into one extraordinary financial event.
Like railways, it demands enormous capital commitments before final demand is known.
Like electricity, it may be a general-purpose technology requiring massive complementary infrastructure before its full productivity effects appear.
Like the internet, its technological importance may be enormous while the eventual distribution of profits remains deeply uncertain, but AI introduces something else, something genuinely strange, the interface looks like software.
Underneath it sits heavy industry, semiconductors, fabrication plants, servers, high-speed networks, data centres, cooling systems, land, power generation, transmission infrastructure, construction, and staggering amounts of capital. The cloud has acquired a factory floor, and the factory floor is filled with chips.
This matters because the AI story is escaping the technology sector, it is becoming an infrastructure story, an electricity story, a construction story, a semiconductor story, a capital-markets story, a debt story and potentially, a macroeconomic story.
Why Now?
Because the numbers are beginning to move from extraordinary to economically significant.
The book cites evidence that global corporate AI investment more than doubled in 2025, while private AI investment increased 127.5% to approximately $344.7 billion. Major cloud providers have simultaneously accelerated spending on computing and AI infrastructure.
More striking still, the book cites the Federal Reserve’s July 2026 Monetary Policy Report: U.S. business fixed investment rose at an 11% annual rate in the first quarter of 2026, with most of the recent strength appearing connected to infrastructure required for AI services.
Read that again.
We are no longer talking only about venture capitalists placing speculative bets on clever software companies.
AI infrastructure is becoming large enough to show up in the machinery of the economy itself, and that changes everything, because capital expenditure generates economic activity before anyone knows whether the eventual investment will produce an adequate return.
Build a data centre and somebody sells the chips. Someone supplies the transformers. Someone constructs the building. Someone installs the cooling. Someone generates the electricity. Someone receives the salary. Someone books the revenue. Someone’s share price rises. Investment made in anticipation of tomorrow’s productivity therefore creates today’s income, and today’s income can then appear to validate yesterday’s optimism.
The book describes this as one of the most subtle mechanisms of the Capital Spiral: expectations can temporarily manufacture some of the evidence used to justify those expectations.
That is not necessarily fraud, it is much more interesting than fraud. It is macroeconomics.
AI’s Most Dangerous Outcome May Be Success
There is another possibility investors should consider, what happens if AI gets dramatically better? The instinctive answer is obvious: AI companies become dramatically more valuable. Maybe.
But the Capital Spiral forces us to ask another question, what if better models, cheaper hardware, greater efficiency, stronger competition and increasingly capable open systems cause the cost of useful machine intelligence to collapse?
Society could win enormously; AI adoption could explode. Entire industries could reorganise around cheap cognition, but owners of expensive AI infrastructure might discover something deeply unpleasant:
Their assets can depreciate without physically deteriorating. The book calls this “depreciation without rust.” The server still works. The data centre still stands. Nothing broke, but somebody else found a cheaper way to produce the same intelligence.
Or a better model appeared, or fewer chips became necessary, or competition crushed pricing power. The asset did not become physically obsolete, its scarcity disappeared.
This creates the extraordinary possibility that the faster AI succeeds technologically, the faster some AI investments could depreciate economically. That is not an argument against AI, it is an argument against confusing technological greatness with investment greatness.
What Happens When Intelligence Gets a Price?
This may be the book’s most radical question. Electricity eventually acquired standard units. Energy could be measured, priced, contracted and financed. What is the equivalent unit for artificial intelligence? Not tokens. Not parameters. Not FLOPS.
Those measure aspects of computation; the economically important question is something closer to:
How much useful cognition can one dollar buy? What does it cost to review a contract? Diagnose a fault? Resolve a customer inquiry? Write a software module? Analyse an experiment? Design a component? Make a decision?
Once machine cognition becomes sufficiently measurable, repeatable and economically productive, something profound happens.
It becomes a factor of production, and once something becomes measurable and economically productive, capitalism does what capitalism has always done. It prices it, contracts it, forecasts it, finances the infrastructure producing it and eventually creates financial claims against its future output.
The book calls this the Financialisation of Intelligence Hypothesis. Railways allowed finance to place claims on future traffic. Utilities allowed claims on future electricity consumption. Consumer finance placed claims against future household income. Mortgage securitisation transformed cash flows into collateral. Crypto experimented with turning financial claims into software.
AI could push the process somewhere entirely new: capital may begin constructing claims on future machine cognition itself. At that point, AI stops being merely a technology story.
It becomes a new chapter in the history of capital.
The Question Isn’t Whether AI Wins
This is perhaps the most important warning in The Future Is Always Overpriced. The world keeps asking: Will AI succeed?
But investors should be asking several different questions. Will AI work? Will it be widely adopted? Who captures the economic value it creates? How intense will competition become? How quickly will the price of intelligence fall? How much infrastructure will be required? Who finances that infrastructure? How much debt will sit behind it?
And finally: How much of AI’s future success has already been paid for today?
Those questions can have radically different answers. A technology can dominate civilisation while its early investors earn terrible returns. A company can create enormous social value while capturing only a fraction of it. Consumers can become vastly richer while capital providers discover that competition destroyed the margins they expected, and the infrastructure built during an investment frenzy can become the foundation of the next economic era precisely because a crash made that infrastructure cheap enough for everyone else to use. That is the Capital Spiral Metaphor.
The Boom Is Not the Revolution
Eventually the AI boom will end, that does not mean AI will fail. Every successful technological boom eventually loses its name. Nobody talks about joining the “electricity revolution” when they switch on a lamp. Nobody announces that they are participating in the “internet revolution” when they send an email. Successful technologies become boring. They disappear into everyday economic life. They become infrastructure.
The book argues that the AI investment boom and the AI economic revolution should therefore never be confused. The boom appears in valuations, venture rounds, semiconductor orders and data-centre construction. The revolution appears somewhere much less glamorous.
Inside the company handling twice as much work with the same workforce. Inside the laboratory testing thousands of hypotheses instead of hundreds. Inside the small business acquiring capabilities previously available only to giant corporations. Inside the worker accomplishing in hours what once required days.
That is when possibility becomes productivity, and history suggests the financial boom can peak before the economic revolution truly begins.
So Why Read This Book Now?
Because we may be living inside the most consequential turn of the Capital Spiral Metaphor yet.
The debate around AI is currently trapped between two tribes. One screams: “This changes everything.”
The other replies: “It’s a bubble.”
The Future Is Always Overpriced argues that history allows for a far more disturbing answer: They could both be right.
AI may genuinely transform civilisation; we may genuinely be constructing enormously valuable infrastructure.
Machine intelligence may genuinely become a new factor of production, productivity may genuinely accelerate, and capital may still build too much, too quickly, at prices that future cash flows cannot justify.
That is what makes this moment dangerous, not because AI is fake, because it may be real enough to convince us that price no longer matters.
Tomorrow Has Already Been Ordered
Capitalism possesses an extraordinary power.
It can take resources from the present and deploy them against a future that does not yet exist, that power built railways before the passengers arrived. Electricity networks before every electrical appliance had been invented. Fibre networks before the traffic existed to fill them.
And today it is building enormous computational infrastructure before anyone can know precisely how much economically valuable intelligence will eventually pass through it. Sometimes this mechanism produces spectacular waste.
Sometimes it produces spectacular progress, history suggests it frequently produces both.
That is why The Future Is Always Overpriced is neither an attack on markets nor a celebration of technological speculation.
It is an attempt to understand the contradiction sitting at the heart of modern capitalism:
We cannot build the future without financing it before we know what it is worth, and once enough money starts moving, belief does something extraordinary.
Belief attracts capital. Capital creates infrastructure. Infrastructure changes reality. Changed reality strengthens belief.
The spiral turns, until eventually there is only one judge left, not the pitch deck, not the stock price, not the AI benchmark, not the billionaire prediction, not the market narrative, but Reality.
The machines must produce. The infrastructure must earn. The productivity must arrive. The cash flows must appear, because however extraordinary the technology becomes, capital eventually asks for its money back, and this time, the bill could be enormous.
The future is being built right now, the future may even be worth everything we have been promised, but that does not mean we haven’t already overpaid for it.
Copyright © 2026 by Bahaa Arnouk. All rights reserved. This article or any portion thereof may not be reproduced or used in any manner whatsoever without the express written permission of the author.
This blog should NOT be read as either an investment, political, legal or a business advice, and it only represents the author’s views (Bahaa Arnouk) and does not represent any other body or organization perspectives, and the author has no liability for any reliance or reference made to it by any third party.