There is a story about technology and employment that, once you hear it, lodges itself permanently in the mind. It is a story about the automated teller machine — the ATM — and the curious fact that a device literally named after the job it was supposed to eliminate did not, in fact, eliminate that job.

The first ATM in the United States was installed at a branch of Chemical Bank on Long Island in 1969. Growth was slow through the nineteen-seventies, but by 1990 there were roughly a hundred thousand machines scattered across the country. By the turn of the millennium, the number had surged past four hundred thousand. In 1975, there were about thirty-one ATMs per million Americans; by 2000, that figure had grown to 1,135 — a thirty-seven-fold increase in just twenty-five years. The economics were irresistible: ATMs carried a large upfront installation cost but operated at a fraction of the ongoing expense of a human employee. If any technology was destined to destroy a category of work, this was it.

And yet the destruction never came. As the economist James Bessen of Boston University documented in his influential research, the number of full-time bank tellers in the United States actually grew alongside the ATM boom — from roughly three hundred thousand in 1970 to over six hundred thousand by the early 2000s. What happened was more subtle than simple replacement. ATMs reduced the number of tellers needed to run a single branch — from about twenty to thirteen between 1988 and 2004 — but they also reduced the cost of opening a branch in the first place. Banks, competing furiously for market share, responded by opening more locations. Urban bank branches increased forty-three percent. Fewer tellers per branch, but far more branches in total. The math worked out in the tellers’ favor.

More important than the arithmetic was the transformation in what tellers actually did. Freed from the low-value drudgery of counting cash and processing checks, they migrated toward relationship banking — advising customers on mortgages, selling financial products, handling the messy, human problems that no machine could resolve. The job changed. Employment didn’t decline. Banks even started hiring more college graduates into teller roles, and there is evidence that wages edged upward. The ATM, in short, was complementary: it automated the routine, elevated the complex, and expanded the overall market for human labor.

This is the version of the story that technologists and policymakers have cited for years as reassurance against automation anxiety. It was the version that J. D. Vance, now Vice President, invoked during his 2024 debate with Tim Walz. It is a good story, and it contains real insight about the difference between automating tasks and eliminating jobs. But it also has a second act that most people leave out.

Beginning around 2010, the number of bank tellers in the United States entered a steep, sustained decline. By 2016, full-time teller employment had fallen to roughly 235,000 — down from 332,000 just six years earlier. By 2022, the figure had cratered to about 164,000. The Bureau of Labor Statistics now projects a further thirteen-percent decline through 2034. This was not a delayed ATM shock; the ATM had reached saturation long before. It was the effect of something else entirely — something that had nothing to do with banking.

It was the iPhone.

Apple introduced the smartphone in 2007. Within a few years, mobile banking applications had rendered the physical branch visit optional for a growing majority of customers. Bank branches in the United States peaked at approximately 99,550 in 2009 and have fallen by more than twenty-one thousand since — a decline of roughly twenty-two percent. When people stopped going to branches, banks stopped needing branches. When branches closed, tellers became redundant. The ATM had automated what tellers did. The iPhone automated the reason people visited tellers. The first technology created a new equilibrium. The second collapsed it.

The distinction is important because it maps onto two fundamentally different theories of how new technologies reshape labor markets. The ATM was task automation — the substitution of a machine for a specific, bounded activity within a broader workflow. Task automation tends to be complementary: it removes the routine, increases productivity, and often expands total employment by reducing costs and growing the addressable market. The iPhone, by contrast, enabled something closer to paradigm automation — the construction of an entirely new infrastructure that bypassed the old workflow altogether. Digital banking did not improve the experience of visiting a branch. It eliminated the need to visit one.

That new infrastructure, it is worth remembering, took decades to assemble. It required fiber-optic networks for connectivity, semiconductor advances for processing power, the invention of the personal computer, the development of the smartphone, and the creation of mobile application platforms and digital payment networks. PayPal, the fintech pioneer, started out literally trying to solve the problem of sending money via email. Amazon, in its earliest days as an e-commerce platform, accepted customer payments by mail-in check. The paradigm shift that ultimately rendered bank tellers obsolete was not a single breakthrough but the slow, compounding accumulation of an entirely new technological substrate — one whose implications took a generation to fully materialize.

This brings us, inevitably, to the question everyone in business, government, and the labor force is now asking: is artificial intelligence the ATM, or is it the iPhone?

Had you asked me this question in early 2023, when ChatGPT had just exploded into public consciousness, my answer would have leaned firmly toward complementary. At that stage, the large language models were useful assistants — capable of drafting a rough memo, debugging a Python script, or summarizing a dense research paper, but fundamentally dependent on a human operator to direct them, evaluate their output, and integrate the results into a broader workflow. They were, in the language of the ATM analogy, automating tasks within existing jobs. They were making people faster, not making people unnecessary.

Two years later, that assessment requires revision.

The change has been driven not just by improvements in raw model capability — though those improvements have been dramatic — but by a qualitative shift in how AI systems are deployed. The era of the chatbot, in which a human typed a question and received an answer, is giving way to the era of the agent, in which AI systems autonomously plan, execute, and complete multi-step workflows with minimal human oversight.

Consider the trajectory. In 2023, using an AI model productively required sustained, skilled interaction — prompt engineering, iterative refinement, careful verification. By late 2025, agentic AI platforms had emerged that could pull files, read documents, write and execute code, communicate across tools, and deliver finished outputs while their human operators stepped away from the keyboard. The shift from “AI as research assistant” to “AI as autonomous worker” is the difference between a calculator and a self-driving car. One augments a human capability; the other performs the function end to end.

The velocity of this transition is visible in the startup ecosystem. At Y Combinator’s Winter 2025 Demo Day, CEO Garry Tan revealed that the batch was growing at ten percent per week in aggregate revenue — a pace he described as unprecedented in early-stage venture history. For roughly a quarter of YC startups, ninety-five percent of their code had been written by AI. Some firms were reaching ten million dollars in annual revenue with teams of fewer than ten people. By the Summer 2025 batch, over sixty percent of YC companies explicitly referenced AI in their pitch, and more than half were building agentic solutions — autonomous systems designed to own entire workflows rather than merely assist with fragments of them.

The most extreme illustration of this new reality may be Medvi, a telehealth startup founded in September 2024 by Matthew Gallagher, a forty-one-year-old entrepreneur from Los Angeles. Gallagher launched the company with twenty thousand dollars in capital and more than a dozen AI tools. He used large language models to write the code for his platform, generate website copy, produce images and videos for advertisements, and handle customer service. He outsourced the regulated medical infrastructure — doctors, pharmacies, shipping, compliance — to partner platforms. In its first full year of operation, Medvi generated four hundred and one million dollars in sales and served more than 250,000 customers, all with a workforce of two people: Gallagher and his brother. The company posted a net profit margin of 16.2 percent — nearly three times the margin of Hims & Hers, a competitor with over 2,400 employees. (It should be noted that Medvi has also attracted scrutiny: the FDA issued a warning letter in early 2026 regarding its compounded GLP-1 products, and the company faces allegations of misleading AI-generated advertisements, including deepfake before-and-after images. The story is remarkable, but it is not uncomplicated.)

The Medvi case is striking not because it is typical but because it was not even conceivable three years ago. Sam Altman, the CEO of OpenAI, had predicted that AI would eventually enable a single person to build a billion-dollar company. When the New York Times profiled Medvi in April 2026, Altman said he had won the bet with his tech CEO friends about when such a company would appear.

Duolingo offers a different but equally telling example. The language-learning company launched its chess course in 2025 — a product conceived and built by two employees who had no prior coding experience and no background in competitive chess. Using AI tools to scaffold lesson logic, write content, and prototype the product, they shipped a course that became Duolingo’s fastest-growing offering, surpassing one million daily active users within months. The course was developed not by domain experts wielding traditional tools, but by novices wielding extraordinarily powerful ones.

If AI were merely an ATM — a tool that automated specific tasks while leaving the broader structure of work intact — the market would be pricing it as a productivity enhancer for existing firms. That is not what the market is doing. In early February 2026, the enterprise software sector experienced what traders called the “SaaSpocalypse” — a two-hundred-and-eighty-five-billion-dollar sell-off triggered by investor fears that AI agents could replace entire categories of commercial software. Within forty-eight hours, ServiceNow dropped seven percent, Salesforce fell seven percent, Intuit plunged eleven percent, Thomson Reuters suffered its worst single-day decline on record, and LegalZoom sank nearly twenty percent. By April 2026, ServiceNow’s stock sat forty-seven percent below its July 2025 peak; Intuit had erased four years of gains, trading at roughly half its 2025 high and posting its worst S&P 500 performance of the new year.

The market’s judgment, in other words, is that AI is not complementary to the existing software ecosystem — it is potentially substitutive of it. Investors are not betting that AI will help ServiceNow sell more seats. They are betting that AI agents may eliminate the need for seats in the first place. The parallel to digital banking is hard to miss: just as the iPhone didn’t improve the branch experience but bypassed the branch entirely, agentic AI threatens not to improve existing SaaS workflows but to bypass the SaaS layer altogether.

This reading is, of course, debatable. Research from MIT Sloan finds that AI’s employment effects depend critically on the breadth of task automation within a given role. When AI automates only some tasks, employment often increases, as workers redirect their efforts toward higher-value activities — precisely the ATM dynamic. When AI automates most tasks within a role, however, firm-level employment in that role drops by approximately fourteen percent. The critical variable is not whether AI can perform a task, but whether, once it performs enough tasks, the remaining human role retains sufficient value to justify the position.

The World Economic Forum has projected that AI and related technologies will generate roughly 170 million new jobs globally by 2030 while displacing 92 million — a net gain of 78 million positions. But aggregate numbers conceal enormous distributional pain. The jobs that are created are not the same jobs that are destroyed, and the people who lose positions in one category are rarely the ones who fill openings in another. A displaced paralegal does not effortlessly become a machine-learning engineer. A laid-off customer service representative does not seamlessly transition into an AI product manager. The economic adjustment, even in optimistic scenarios, will be wrenching for millions of workers — and unlike previous technological transitions, it may unfold at a pace that outstrips the capacity of educational institutions and labor markets to adapt.

What distinguishes the current moment from earlier waves of automation is the speed of infrastructure buildout. The digital revolution that eventually displaced bank tellers required decades of sequential development: fiber-optic cables, then semiconductors, then personal computers, then smartphones, then mobile applications. Each layer depended on the one before it, and the full stack took the better part of forty years to assemble. The AI infrastructure, by contrast, is being constructed with trillions of dollars of concurrent investment in a compressed timeframe. The foundational models are already powerful and improving rapidly. The agentic orchestration layer is emerging in real time. The application layer — specific tools and workflows built atop agents — is proliferating faster than anyone predicted.

This acceleration means that the interval between “AI as productivity tool” and “AI as autonomous worker” is likely to be years, not decades. The ATM-to-iPhone transition in banking played out across a generation. The analogous transition in AI may play out across a business cycle. For firms, the implication is that workforce strategies premised on gradual, incremental adaptation — the assumption that employees will have time to learn new skills, that organizational structures will evolve slowly, that the competitive landscape will shift predictably — may badly underestimate the pace of change.

For policymakers, the challenge is equally urgent. The complementary phase of AI — the ATM phase — is the window in which to invest in retraining programs, educational reform, and social safety nets. Once AI shifts decisively from complementary to substitutive — from automating tasks within jobs to eliminating the rationale for those jobs — the political and social costs become far harder to manage. The precedent of bank tellers is instructive. Nobody organized massive retraining programs for displaced tellers because the decline happened gradually, across years, and was absorbed by the broader labor market. But the scale of potential AI-driven displacement is orders of magnitude larger, and the pace is orders of magnitude faster.

The honest answer to the question — is AI the ATM or the iPhone? — is that it is both, simultaneously, in different domains and on different timescales. For some categories of work, AI will function as a powerful complement: it will automate the routine, elevate human judgment, and expand the overall market for skilled labor. For others, it will enable entirely new business architectures that bypass existing workflows altogether — architectures in which a two-person startup generates hundreds of millions in revenue, in which a language-learning app ships a chess course built by non-experts, in which software companies that dominated for decades see their market capitalizations halve in months.

The paradigm shift is not coming. It is here. The ATMs gave bank tellers a generation to adapt. AI may not be so generous with the rest of us.

Most people reserve their admiration for rock stars or rocket scientists. Mine belongs to a seamstress. Coco Chanel — orphaned at eleven, raised among nuns, schooled in the austere geometry of need — went on to build one of the most recognisable brands on earth. That trajectory, from threadbare beginnings to the summit of Parisian elegance, has always struck me as the most compelling story in fashion.

Chanel’s genius lay in a deceptively simple conviction: that luxury ought to serve the woman wearing it, not the other way around. The little black dress, the collarless tweed suit, the liberation from corsetry, each was an act of quiet rebellion dressed up as good taste. With the financial backing of wealthy patrons, she opened her first shop in Paris in 1910, selling millinery under the name Chanel Modes. What followed was a career spent stripping fashion of its ornamentation and replacing it with something rarer — functionality in service of grace.

It is bittersweet, then, to observe what the luxury industry she helped pioneer has become.

In the wake of the Covid-19 pandemic, a peculiar frenzy gripped the consumer economy. Flush with stimulus cheques and savings accumulated during months of enforced idleness, shoppers embarked on what became known as “revenge spending,” a cathartic binge that disproportionately favoured luxury brands. For a brief, dizzying moment, Bernard Arnault, chairman of the luxury conglomerate LVMH, ascended to the title of the world’s richest man. The champagne, it seemed, would never stop flowing.

But much of the industry’s windfall came not from selling more, but from charging more. The numbers are staggering. Prada’s Galleria Saffiano bags rose 111% between October 2019 and April 2024. Louis Vuitton’s canvas Speedy bags doubled in the same window. These were not incremental adjustments for inflation; they were audacious bets that desire would outrun arithmetic.

For a time, it did. Then it didn’t.

As consumers in the world’s two largest economies, the United States and China, have confronted economic headwinds, demand for luxury goods has fallen sharply. LVMH’s shares have dropped as much as 40% from their 2023 peak and continue to face downward pressure. Burberry and Dior have seen their valuations halved. The steepest decline, however, belongs to Kering, the parent company of Gucci, where industry-wide malaise has been compounded by the brand’s own faltering relationship with its customer base. It turns out that even Veblen goods, those paradoxical products whose allure is supposed to intensify as their prices climb, have a breaking point.

The economic theory behind luxury pricing is elegant in its counterintuitiveness. Thorstein Veblen posited that for certain goods, rising prices thin the ranks of ownership and thereby amplify desirability. A Chanel bag at twice the price is not merely more expensive; it is, in theory, more exclusive, and therefore more coveted. Just imagine if Walmart had doubled the price of everything on its shelves within five years — there would be riots. But luxury operates by a different logic, one in which higher prices are not a deterrent but an invitation.

This logic held for decades, but the recent price escalation has exposed its fragility. The issue, as Katherine Zarrella argued in her New York Times piece on the subject, is not that prices rose, but that they rose without any corresponding improvement in quality or service. Stitching grew looser. Hardware tarnished faster. Customer service became perfunctory. Consumers with deep pockets, the very clientele luxury brands cannot afford to alienate, began asking the most dangerous question a luxury house can face: is this actually worth it?

And there is a more fundamental problem. The Veblen principle functions on scarcity, yet there is nothing scarce about a Louis Vuitton monogram bag when you encounter one every other block in Manhattan or Shanghai. When ubiquity replaces exclusivity, the justification for doubling a price tag within two years evaporates, especially when the target demographic is not the billionaire class, but the aspirational middle class.

Against this backdrop of industry-wide decline, one house has not merely survived but flourished. Hermès posted a 15% sales increase in 2025, even as the S&P Global Luxury Index fell more than 20% the year prior. The answer lies in discipline. Where competitors chased volume, Hermès guarded its supply chain with near-obsessive control. Price increases have been measured, roughly 6–7% per year, rather than the explosive surges favoured by rivals. And the company has remained ruthlessly selective about who can buy its most coveted products. A Birkin bag is not merely expensive; it is genuinely difficult to obtain. The scarcity is not a marketing posture. It is architectural. It is also worth noting that Hermès remains majority-controlled by its founding family, a governance structure that insulates the brand from the quarterly earnings pressures that have driven so many competitors into short-termist excess.

For the consumers priced out, or, perhaps more accurately, who have priced themselves out of caring, alternatives have emerged with surprising vigour. Mid-range brands like Coach and Ralph Lauren have enjoyed a striking resurgence, particularly among younger demographics drawn to their balance of heritage and accessibility. Todd Snyder, the menswear label, has grown from $2 million in sales in 2015 to $130 million in 2024, offering a proposition that luxury houses once claimed as their own: genuine quality at a fair price. Then there is the second-hand market. Platforms like The RealReal and Depop have surged in popularity, offering a secondary avenue to luxury that sidesteps the industry’s price inflation entirely. There is a certain irony in this: the very products that brands inflated beyond reason now circulate in resale ecosystems that undercut the new-product market.

Writing about the luxury industry’s predicament carries a tinge of melancholy, particularly for anyone who admires the craft and vision that once defined these houses. The pivot towards short-term earnings, aggressive pricing, expanded product lines, ubiquitous distribution, has eroded the very identity that allowed these brands to command devotion over decades. This applies beyond apparel. Parallel dynamics are visible in luxury automobiles, with Porsche’s diluted exclusivity, and in high-end hospitality, where prices have become absolutely insane.

Zarrella lamented that today, instant gratification and appearances have become more desirable than substance or intrinsic worth. She may be right about the mainstream luxury houses. But the death of conspicuous, logo-heavy luxury has quietly made room for something more interesting. Brunello Cucinelli, whose founder espouses a humanistic philosophy rooted in community and craftsmanship, has thrived precisely because its identity feels earned rather than manufactured. Its cashmere is not merely expensive, it is traceable and dyed and sewn by artisans paid above-market wages.

Oscar Wilde once quipped that people know the price of everything and the value of nothing. He was writing about Victorian society, but he might as well have been diagnosing the modern luxury industry. Perhaps the most useful legacy of its current reckoning is this: a reminder that scarcity and value are not strictly functions of price. Coco Chanel understood this intuitively. She built a brand on the radical premise that elegance should be affordable, functional, and honest. The industry she helped create forgot that lesson. Some, at least, appear to be remembering it.

There is a conventional wisdom, that if you play stupid games you will win stupid prizes. Three weeks into Operation Epic Fury, that aphorism has acquired the weight of prophecy. What began on February 28th as a joint American-Israeli aerial assault, ostensibly to prevent Iran from acquiring nuclear weapons, has spiralled into the most expensive and strategically incoherent American military operation since the invasion of Iraq in 2003. Brent crude has surged past $109 a barrel. The Strait of Hormuz, through which a fifth of the world’s oil passes, is effectively closed. Thirteen American service members are dead, over 1,300 Iranians have been killed including Supreme Leader Khamenei, and the Pentagon has spent upwards of $16.5 billion—and is now requesting $200 billion more from Congress.

Yet the gravest cost may not be measured in barrels or billions. It is the revelation that America’s commander-in-chief has no coherent idea of what America’s role in the world should be, and no foreign policy worthy of the name. The Iran war has not created this crisis. It has merely made it impossible to ignore.

The most damning indictment of Mr Trump’s leadership is the simplest, that nobody—including, it appears, the president himself—can say what the war is for. Officials have variously stated the objective is to destroy Iran’s nuclear programme, eliminate its missiles, degrade its proxies, secure its resources, punish the regime for killing protesters, and bring about regime change. Vice-President Vance told reporters “we are not at war with Iran; we’re at war with Iran’s nuclear programme.” Hours later, Mr Trump posted on Truth Social calling for regime change. Karim Sadjadpour of the Carnegie Endowment captured it well: the president has been “all over the place” on his aims, and that lack of clarity has put the military and America’s partners in impossible positions. Asked on March 16th whether he was ready to declare victory, Mr Trump said no. He has separately said the war will end when he “feels it in his bones.”

The incoherence is dangerous. A war begun as what Mr Sadjadpour calls a “war of choice”—no imminent nuclear threat, no impending attack on American assets—has morphed into a “war of necessity” now that Iran has closed the Strait of Hormuz. American intelligence assessments contradict the administration’s claims of an immediate Iranian missile threat, placing such capabilities as far off as 2035. The IAEA confirmed it had no evidence of a structured weapons programme at the time of the strikes. The casus belli rests on presidential intuition, which the White House press secretary has defended as a “feeling based on fact.”

Mr Trump’s failure is rooted in a deeper intellectual deficiency. He has never grasped that foreign policy involves more than leverage and deal-making. He treats alliances as protection rackets, trade agreements as zero-sum contests, and diplomacy as the art of extraction. This framework has a brutal logic in tariff negotiations. Applied to the Middle East, it is catastrophic. Iran is a nation of 88 million people with a theological state, a 3,000-year civilisational identity, and a resistance to foreign domination forged by decades of Western interference—from the CIA’s coup against Mosaddegh in 1953 to American backing of Saddam Hussein’s war against Iran in the 1980s. Mr Trump appears genuinely baffled that Iran has not capitulated. On March 17th he claimed “nobody” expected Iran to retaliate against Gulf states, even though Iranian officials had publicly vowed to do precisely that.

If Mr Trump’s war aims are confused, his alliance management has been farcical. Having spent years insulting NATO partners, threatening to seize Greenland, and imposing tariffs on allies, he turned in mid-March to those same countries and demanded they send warships to the Strait of Hormuz. The response was devastating. Germany’s defence minister asked what Mr Trump expected “a handful of European frigates to do that the powerful US Navy cannot,” adding: “This is not our war; we have not started it.” France said it would “never take part” in operations to reopen the strait. Spain, Italy, Australia, Japan, Poland, Sweden, and Greece all declined. The EU’s foreign-policy chief said there was “no appetite” to expand naval operations. Estonia’s foreign minister posed the question haunting every European capital: “What will be the plan?”

Mr Trump’s reaction followed a familiar script. On Monday he claimed “numerous countries” were on their way to help, but refused to name any. By Tuesday, after virtually every invited nation had publicly refused, he reversed course: “We don’t need any help, actually.” The volte-face would be comic were its consequences not so severe. The United States is prosecuting a major war in near-total diplomatic isolation, a state of affairs unprecedented since at least the Second World War.

The financial toll is unforgiving. The Centre for Strategic and International Studies estimated costs at $891 million per day during the opening phase. Each Patriot interceptor costs roughly $4 million; Iran launched over 2,000 drones and 500 ballistic missiles, producing a cost ratio of 106-to-1 between interceptor and drone. Petrol prices have surged to $3.88 per gallon, nearly a dollar more than a month ago. Mr Trump has waived the Jones Act and tapped the Strategic Petroleum Reserve, but analysts say the impact on prices will be negligible. Senator Elizabeth Warren summarised the political problem, that while there is no money for the 15 million Americans who lost health coverage, there is a billion dollars a day for bombing Iran.

Beneath the chaos lies a question Mr Trump has made urgently pressing: what is American foreign policy for? For eight decades the answer was reasonably clear—a liberal international order underpinned by free trade, collective security, and multilateral institutions. Mr Trump has dismantled that framework without erecting anything in its place. He withdrew from the Trans-Pacific Partnership, the Paris accord, and the Iran nuclear deal. He imposed tariffs indiscriminately and publicly mused about not defending Taiwan. His “America First” slogan implies retrenchment and restraint. But an America First president does not launch a war of choice in the Middle East without consulting allies, without an exit strategy, and without stable war aims. The contradiction at the heart of Trumpian foreign policy is that it is simultaneously isolationist and aggressive, parsimonious and profligate, contemptuous of allies and desperate for their help. It is, as one analyst put it, “instinct, not strategy.”

History offers ample warning. Every American president who has taken the country into a protracted conflict without clear objectives has paid a devastating political price—Truman in Korea, Johnson in Vietnam, Bush in Iraq. Mr Trump’s net approval has fallen to -15.3, a second-term low. A Quinnipiac poll found just 37 per cent approving of his performance. Among true independents, only 24 per cent approve. There has been no rally-round-the-flag effect; there is only exhaustion, the accumulated residue of two decades in Afghanistan and Iraq. Joe Kent, a former counterterrorism director in Mr Trump’s own administration, has resigned, arguing Iran poses no imminent threat.

Mr Trump has always loved chaos and thrived in it. But this is not a tariff that can be imposed and withdrawn at whim. It is a war, with all the irreversibility that wars entail. The Strait is closed. Thirteen Americans are dead. The bill approaches $200 billion. The allies are gone. And the president still cannot say what he is fighting for. A superpower that does not know what it wants is more hazardous than one that wants the wrong things. At least the latter can be negotiated with and contained. The former is simply unpredictable, and in a world of nuclear weapons and interconnected economies, unpredictability is the gravest threat of all.

Imagine you run a mid-sized software company. You need capital to grow, but the banks say no — your business is too risky, too small, or doesn’t fit neatly into their lending criteria. Where do you go?

Increasingly, the answer has been private credit: investment funds that step in where traditional banks won’t, offering loans to businesses that fall outside the conventional banking system. Over the past decade, private credit has grown from a niche corner of finance into a multi-trillion-dollar industry, filling a gap that banks left open after tightening their lending standards in the wake of the 2008 financial crisis. This matters beyond Wall Street. Private credit funds the companies that employ people, build infrastructure, and develop technology. When private credit is healthy, capital flows to businesses that need it. When it isn’t, the effects ripple outward — into jobs, into pensions, and into the broader economy.

Last summer, private credit was the talk of the financial world. The explosion of investment in artificial intelligence had created an enormous appetite for capital — companies needed billions to build data centers, develop models, and scale infrastructure. One of the marquee deals was Blue Owl Capital’s $27 billion arrangement with Meta to finance data center construction in Louisiana. It was the kind of headline that signaled private credit had arrived as a dominant force in modern finance.

But even at the peak, there were warning signs. Conversations with investment professionals at the time revealed a shared unease: the space was getting crowded. More funds were entering, competition for deals was intensifying, and the pressure to deploy capital meant that underwriting standards — the rigor with which lenders evaluate borrowers — were beginning to slip. The consensus was that “dispersion” would define the industry going forward. Some funds would thrive, but others, the ones that cut corners to win deals, would eventually pay the price.

The test came sooner than anyone expected, and ironically, it came from AI itself — the very force that was supposed to propel the industry to new heights. Earlier this year, Anthropic’s release of Cowork and its suite of AI agents sent a shockwave through financial markets. The concern was straightforward but devastating. If AI agents could automate much of what traditional software companies sell — customer support, data analysis, workflow management — then the business models underpinning thousands of software firms were suddenly in question. These companies largely rely on selling subscriptions on a per-seat basis: the more employees a client has, the more they pay. AI threatens to replace that model with one based on token consumption, where a single AI agent could do the work of many seats at a fraction of the cost.

The public markets reacted swiftly. The IGV software ETF fell more than 15% year to date. High-quality companies like Intuit dropped over 30%, not because their businesses had collapsed, but because investors feared they would. The second-order effects hit private credit hard. Many of the companies in private credit portfolios are exactly the kinds of mid-market software firms most vulnerable to this disruption. And because private credit funds don’t trade on public exchanges the way stocks do, the stress didn’t show up immediately in prices. It showed up in something more dangerous: redemptions.

A Slow-Motion Bank Run

To understand why redemptions are so dangerous for private credit, you need to understand how these funds value their assets, and why that process is fundamentally different from public markets.

Here’s a simplified example. Say a private credit fund lends $100 to a software company for five years at 8% interest. The fund records this loan as a $100 asset on its books. The next day, fears about AI disruption cause the market to demand higher interest rates from similar software companies — say, 8.5% instead of 8%. What is that loan now worth?

In public markets, the answer is clear, roughly $98. When market interest rates go up, the value of existing lower-rate debt goes down. If you sold the note today, the buyer would need a discount to achieve the new market rate of 8.5%. This is called “mark-to-market” — valuing your assets at what the market would actually pay for them right now.

But private credit funds don’t have to play by these rules. A fund manager might reasonably argue: “Nothing has changed with the borrower. They’re still making payments. I’m not selling the loan. It will pay $100 at maturity plus interest. Why would I mark it down to $98 based on market sentiment that might reverse tomorrow?” This is called “mark-to-model” — valuing assets based on an internal assessment rather than market prices. Both approaches have legitimate uses. But the tension between them creates a dangerous vulnerability.

Many of the largest private credit firms run retail-facing vehicles called business development companies (BDCs). Some of these BDCs allow investors to redeem their shares — not whenever they want, but in limited amounts, typically once per quarter. Crucially, these redemptions happen at net asset value, which is based on the fund’s own marks. If a fund says its loans are worth $100 but the market believes they’re worth $98, investors have a powerful incentive to redeem. They can cash out at $100 for something they believe is worth only $98, a built-in arbitrage. And once some investors start redeeming, others follow, fearing they’ll be left holding assets that are worth even less once the selling pressure hits.

This is the dynamic that has been playing out across the private credit industry in recent months. Redemption requests have surged to historically high levels. Some funds have been forced to limit withdrawals, creating scenes uncomfortably reminiscent of the Silicon Valley Bank run — except this time it’s happening in private markets, where there’s less transparency and fewer safety nets. The very act of restricting redemptions can accelerate the panic. When investors hear that a fund is gating withdrawals, it confirms their suspicion that the assets aren’t worth what the fund claims. Trust erodes, and the cycle intensifies.

Leverage makes all of this worse. Many private credit funds borrow money from banks to boost their returns, using their loan portfolios as collateral. This works well when asset values are stable. But when values start to fall — or when banks suspect they might — lenders can issue margin calls, demanding that funds post additional collateral or pay down their borrowings. To meet those calls or fund redemptions, managers may need to sell assets. But selling illiquid private loans at a discount further depresses valuations, which triggers more margin calls and more redemptions. It’s the same doom loop that has brought down financial institutions throughout history, from Long-Term Capital Management in 1998 to Bear Stearns in 2008.

Why This Matters Beyond Wall Street

The private credit crunch is not happening in isolation. It follows the high-profile bankruptcy of subprime auto lender First Brands and the collapse of auto-parts company Tricolor — losses significant enough that investors have filed lawsuits against Jefferies, which was involved in lending to First Brands. These events are testing an industry that has enjoyed an almost uninterrupted run of growth since the financial crisis.

The implications extend well beyond the investors in these funds. For the broader economy, private credit is now a major source of financing for mid-sized businesses. If funds become more cautious or pull back lending, companies that rely on this capital — for payroll, for expansion, for operations — could find themselves squeezed. That translates into slower growth, hiring freezes, or layoffs.

For retirement savings, many pension funds and insurance companies have allocated heavily to private credit in search of higher returns. If these investments underperform, it could affect the retirement security of millions of people who have never heard of a BDC. And for the financial system at large, the interconnection between private credit funds and traditional banks through leverage creates channels for contagion. Stress in private credit doesn’t stay contained in private credit.

This isn’t the end of the industry. Private credit serves a genuine economic need — channeling capital to borrowers that banks can’t or won’t serve — and that need isn’t going away. But the current turbulence will reshape the landscape. The funds that underwrite carefully, maintain conservative leverage, and communicate honestly with investors about their marks will emerge stronger. Those that stretched for yield, overleveraged, or relied on opaque valuations to mask deteriorating portfolios will not. The “dispersion” that industry insiders warned about last summer is no longer a prediction. It’s happening now.

The lesson for investors and observers alike is an old one, made new by modern finance: when the price of something is hard to see, it’s easy to pretend it hasn’t changed. But pretending doesn’t make it so — and eventually, reality catches up.

There is a running joke on the internet that President Trump has a calendar of foreign leaders he intends to take down each month in the lead-up to the midterms—almost comical, as if he is collecting Pokémon. In January, the United States through a swift military operation captured Maduro, the leader of Venezuela. In February, the U.S. assisted the Mexican government in eliminating the cartel leader known as “El Mencho.” And as March arrived, President Trump launched what is effectively a war under Operation Epic Fury, killing the Iranian Supreme Leader Ali Khamenei during the opening wave of airstrikes.

With each operation, the magnitude of response has escalated. The capture of Maduro was drama-free and broadly popular. The killing of El Mencho sparked a sharper backlash, as cartels attacked airports and burned vehicles across parts of Mexico, stranding tourists and cancelling flights. The most significant response, by far, has come from Iran, which launched hundreds of ballistic missiles and drones against Gulf states including the United Arab Emirates, Kuwait, Qatar, and Saudi Arabia, and effectively shut down the Strait of Hormuz, through which roughly 20% of the world’s oil and gas supply passes.

While at the time of this writing, Mr. Trump has signaled that the war may be reaching its conclusion, telling CBS News the conflict is “very complete, pretty much”—the world is much different than it was before a week of fighting between the U.S. and Israel against Iran. To explore the implications, I want to play a game called Two Losers and a Winner, inspired by the classic icebreaker of two lies and a truth.

The First Loser: Energy-Dependent Nations

The first and perhaps the most glaring takeaway is that hydrocarbons remain foundational as an energy source for nearly every country on earth. Oil and gas power industries from manufacturing to transportation, agriculture to electricity generation. To become economically secure, a country must first consider how to become energy secure, much as the Covid-19 pandemic taught the world the importance of securing domestic supply chains for critical industries. Without energy security, countries remain hostage to a fragile global supply chain, vulnerable at a handful of strategic chokepoints.

The Strait of Hormuz is the most consequential of those chokepoints. It is a narrow waterway just 21 miles wide at its narrowest point, separating Iran and Oman, through which approximately 20 million barrels of oil pass each day, roughly one-fifth of global seaborne oil trade. An estimated 84% of crude shipments through the strait are destined for Asian markets, with China, India, Japan, and South Korea accounting for nearly 70% of those shipments, according to the U.S. Energy Information Administration. Europe, too, depends on the strait, as roughly 30% of the continent’s jet fuel supply originates from or transits through it, and about 12–14% of Europe’s liquefied natural gas (LNG) comes from Qatar via this corridor.

The conflict exposed this dependency with brutal clarity. Within days of Operation Epic Fury, tanker traffic through the strait collapsed by roughly 70% before falling to virtually zero, as the Islamic Revolutionary Guard Corps declared the strait closed and threatened to set ablaze any vessel attempting passage. Over 150 ships anchored outside the strait, unwilling to enter. Insurance premiums for very large oil tankers surged by hundreds of thousands of dollars per transit. Brent crude, which traded around $70–73 per barrel before the war, spiked above $100 per barrel for the first time since Russia’s 2022 invasion of Ukraine, hitting an intraday high near $120 before settling around $99. U.S. gasoline prices jumped roughly 17% within a week of the war’s start, with analysts warning of $4–45 per gallon if the strait remained closed.

The LNG market was hit just as hard. On March 2, Qatar, the world’s largest LNG exporter, announced it had halted all gas production after Iranian drones struck Qatari gas facilities, and declared force majeure on its gas contracts two days later. European natural gas prices nearly doubled in the aftermath. Goldman Sachs Research estimated that a hypothetical two-month disruption of LNG supply through the strait could push European natural gas prices above €100 per megawatt-hour, compared to roughly €32 before the war. Qatar’s energy minister warned that if the war continued, other Gulf producers could be forced to halt exports, which “will bring down economies of the world.”

The downstream effects on developing nations are catastrophic. I witnessed this dynamic firsthand in Sri Lanka in 2022, in the wake of the Covid-19 pandemic. The country’s rampant fiscal expansion had led to surging inflation and economic stagnation. One of the core drivers was the inability to import energy at affordable prices. As the Sri Lankan rupee collapsed, the cost of fuel imports soared, dragging down agriculture, transportation, and manufacturing. During my visit to the urban slums of Colombo, many workers were unable to commute because the cost of transportation had become unaffordable. Families struggled to make ends meet, and many turned to crime—theft, robbery, and even trafficking of crystal methamphetamine. Trust within communities eroded. Children were pulled from schools, unable to afford basic supplies. These consequences set the nation back by decades.

The same dynamics now threaten countries across Asia. Japan’s refiners, who source roughly 95% of their crude from the Gulf, have already asked the government to release stockpiled oil. Pakistan, which sources 99% of its LNG imports from Qatar, faces an acute crisis. India and China, while more diversified, are still deeply exposed. The IMF has warned that if oil prices remain above $100 per barrel for a sustained period, it could shave 0.4 percentage points off global GDP growth and add 1.2 percentage points to inflation.

The world’s dependence on hydrocarbons, particularly in emerging markets, is often glossed over as media outlets advocate for cleaner alternatives like wind, solar, and nuclear. While diversifying energy sources is essential, this conflict has again demonstrated that the world remains deeply reliant on oil and gas—and that for developing nations, a lack of access to cheap energy is not merely an inconvenience but a direct path to economic and social turmoil.

The Second Loser: The Gulf States

For the past decade, Gulf states, particularly the UAE and Qatar, have invested enormous effort in transforming their economies away from a singular dependence on hydrocarbons. Dubai has become synonymous with luxury tourism, financial services, and architectural ambition. Qatar hosted the 2022 FIFA World Cup. Saudi Arabia has poured billions into entertainment, sports, and its Vision 2030 diversification agenda. More recently, the UAE has positioned itself as a hub for artificial intelligence and technology. The pitch to the world was clear: the Gulf is stable, modern, and open for business.

The war with Iran shattered that narrative. Within hours of Operation Epic Fury, Iran launched retaliatory strikes across the entire Gulf Cooperation Council. The UAE alone reported facing 174 ballistic missiles, over 680 drones, and eight cruise missiles in the initial waves. While the vast majority were intercepted, the debris and the projectiles that got through inflicted real damage. Dubai International Airport, the busiest airport in the world for international traffic, was struck and temporarily shut down, stranding roughly 20,000 passengers in the UAE alone and another 8,000 in Qatar. A drone struck near the Fairmont Hotel on Palm Jumeirah, one of Dubai’s most iconic luxury destinations. The Burj Al Arab and Jebel Ali Port were damaged by falling intercept debris. An Amazon Web Services data center caught fire after being hit. Four people were killed in the UAE, foreign workers from Pakistan, Nepal, and Bangladesh, and over 100 were injured.

The images of smoke rising over Palm Jumeirah and yachts sailing past burning port facilities struck at the heart of the Gulf’s brand. As one analyst observed, seeing Dubai and Doha bombed is as jarring for the region as seeing Miami or Seattle bombed would be for Americans. For a region that has attracted the world’s wealthy by projecting safety and stability, this was a fundamental breach of trust. In the days following the strikes, banks reportedly began offering employees relocation packages away from the region. Airlines including Emirates, Etihad, and Qatar Airways, which together operate over a thousand flights daily, suspended operations entirely.

This is not to say that companies will abandon the region altogether. The Gulf’s infrastructure, tax advantages, and geographic positioning remain compelling. But the war has forced businesses, investors, and wealthy individuals to price in geopolitical risk in a way they previously did not. The plan for the region to become a global hub for entertainment, finance, and technology may no longer follow the trajectory its leaders envisioned. Businesses will be more cautious about long-term capital commitments. Tourists and expatriates will weigh safety considerations they once took for granted. While the Gulf states were largely spared during the tit-for-tat exchanges between Iran and Israel in 2024 and 2025, this time a sustained week of fighting has inflicted damage that may prove economically irreversible in the short term.

The Winner: Wall Street

Amid all this destruction and disruption, there is a winner—and, remarkably, it is Wall Street. Once again, markets called Mr. Trump’s bluff and, so far, have largely gotten it right. Despite oil surging past $100, despite a region in flames, the S&P 500 has displayed surprising resilience. On the first trading day of the war, the Dow fell nearly 600 points intraday before recovering to close down just 73 points, while the S&P 500 and Nasdaq actually finished in the green. Even on Monday, March 9, after oil briefly spiked near $120 a barrel, the S&P 500 rebounded from an intraday loss of over 1.5% to close up 0.83% at 6,796 after Trump signaled the war’s end. The Dow, despite its worst weekly decline since April, still managed to recover 239 points on the day.

There are structural reasons for this resilience. The S&P 500 is now heavily concentrated in mega-cap technology and AI companies that are far less exposed to energy prices and physical supply chain disruptions than the manufacturing-heavy index of decades past. Research from Carson Group, which compiled data from 40 major geopolitical events over the past 85 years, found that the S&P 500 has historically lost just 0.9% in the first month following a major geopolitical shock but gained 3.4% over the following six months. Investors also had time to position for this conflict—the military buildup had been well-telegraphed, and Wall Street widely expected a contained, short-duration operation.

This market behavior has become a recognizable pattern under the Trump administration. Since “Liberation Day” in April 2025, when sweeping tariff announcements sent the market down over 10% within weeks, investors have learned to treat each new escalation—threats over Greenland, additional tariffs, now a war—with a degree of skepticism. The market has essentially adopted a “wait and see” posture, betting that disruptions will be temporary and that the administration will de-escalate before permanent economic damage is done.

Yet there are limits to this resilience. The week ending March 7 was the worst combined week for stocks and bonds since the tariff shock last April, as both fell simultaneously, the inflationary shock of an oil supply disruption pushed Treasury yields higher rather than lower, breaking the traditional crisis playbook. JPMorgan’s trading desk turned “tactically bearish” on Monday, warning that a 10% correction from the S&P 500’s peak, down to roughly 6,270, remained possible if the conflict dragged on. The $100-per-barrel oil threshold is widely viewed on Wall Street as the level at which real economic damage begins to compound.

Still, the market’s ability to absorb a hot war with Iran without a major selloff is a testament to the structural strength of American companies—and also, perhaps, a reminder of the limited extent to which the U.S. equity market reflects the broader American economy or the global economy. The S&P 500 is not measuring the pain of stranded tanker crews, Pakistani workers killed by debris in Abu Dhabi, or Japanese refiners scrambling for crude. It is measuring the profit expectations of Nvidia, Apple, and Microsoft. Whether that disconnect is a feature or a flaw depends on your perspective.

Will Next Time Be Different?

In the end, the war in Iran has demonstrated just how fragile and interdependent the global order remains. Geopolitical shocks produce winners and losers, and the distribution of those outcomes is rarely equitable. The conflict in Iran threatens regional stability across Asia, where social unrest is already common, and could derail the Gulf’s ambitious economic transformation. Sustained high oil prices risk stagflation in the United States and could tip developing economies into outright crisis.

As the conflict gradually de-escalates and the first tankers resume passage through the Strait of Hormuz, it is worth asking a simple but uncomfortable question: will next time be different? The answer, almost certainly, is no. So long as the world remains dependent on a handful of narrow waterways for its energy supply, and so long as geopolitical rivalries persist, shocks like these will recur. The only variable is whether countries, and markets, will be better prepared when they do.

Every major technological wave triggers the same debate: is this a tool that augments human labor, or a force that displaces it? For most of industrial history, the answer has been the former. Machines made workers more productive; they rarely made workers obsolete. AI is beginning to look like an exception to that rule, and understanding why requires a clear-eyed look at what makes this wave structurally different from the ones that came before.

As investor Howard Marks observed in a recent memo, the improvements in AI have shifted from incremental to transformational, from a tool that enhances human productivity to one capable of operating fully autonomously, replacing rather than merely assisting the low- to mid-level knowledge worker. That distinction has profound implications for the labor market, for capital allocation, and for where economic value accrues across industries.

The clearest early evidence of this shift is autonomous driving. Waymo’s driverless vehicles now operate commercially across San Francisco, Los Angeles, Austin, and Washington D.C., logging millions of rider-only miles with significantly fewer injury-causing crashes per mile than human drivers. This is not an incremental safety upgrade layered on top of a human driver. There is no human driver. The economics of transportation, long built around the cost of labor, are being restructured from the ground up. But autonomous driving is a narrow domain. The broader and more consequential question is what happens when this logic of full autonomy is applied to knowledge work — to law, finance, accounting, software engineering, and analysis. To answer that, history offers a useful mirror.

Few inventions illustrate the mechanics of technological disruption as sharply as the cotton gin, patented by Eli Whitney in 1793. To understand its impact, you have to understand the bottleneck it solved. In the late eighteenth century, short-staple cotton — the variety that grows readily across the American South — was commercially worthless despite high global demand. The fibers of short-staple cotton cling tightly to their seeds, and separating them by hand was brutally slow. A single worker could clean roughly one pound of cotton per day. At that productivity rate, the economics of growing cotton in the American interior simply did not work.

Whitney’s cotton gin changed that overnight. A hand-cranked gin could process roughly 50 pounds of cleaned cotton per day, a 50-fold increase in throughput. The effect on output was staggering: U.S. cotton production exploded from roughly 1.5 million pounds in 1790 to over 35 million pounds by 1800, and to nearly 2 billion pounds by the eve of the Civil War in 1860. By the 1850s, the American South was producing roughly two-thirds of the world’s cotton supply, and cotton accounted for more than half of all U.S. export revenue. The South had not merely grown a crop, it had become the beating heart of the global textile industry. King Cotton was not a nickname; it was an economic reality.

But the cotton gin’s most instructive feature is what it did not do. It automated only one stage of the production chain, the separation of seed from fiber. It could not pick cotton from the field. That process remained entirely manual, and the explosive profitability unlocked by the gin made planters desperate to scale picking capacity as fast as possible. The result was a catastrophic expansion of slavery: the enslaved population in the South grew from approximately 700,000 in 1790 to nearly 4 million by 1860. The gin did not reduce the demand for human labor in aggregate, it concentrated and brutalized it in the one stage of production it could not touch.

Over the following century, mechanization completed what Whitney started. Mechanical cotton pickers, developed commercially in the 1940s and 1950s, could harvest as much in a day as 50 hand-pickers. By the 1970s, virtually all American cotton was machine-harvested, and what had once required an enslaved labor force of millions was reduced to a capital-intensive operation run by a handful of operators. The economics of production were fully commoditized; the value migrated upstream. Today, you can buy a plain cotton t-shirt for a few dollars. The production itself is worth almost nothing.

Where the value now lives is in brand, design, and the cultural meaning a company wraps around its products, in deciding what problem to solve and for whom. Nike offers the sharpest example. If you trace the supply chain, the workers stitching together sneakers in Vietnam and Indonesia are paid very little, and their wages are under constant competitive pressure from lower-cost geographies.

The company spends over $1.5 billion annually on research and design, and the vast majority of the value it captures flows from brand equity built across decades of cultural marketing. Remove the Swoosh, and you can purchase a nearly identical shoe, the growing “dupe” market does exactly this, for a fraction of the retail price. Skechers sells comparable footwear at a steep discount for the simple reason that its brand carries nowhere near the cultural weight. The production method is a commodity. The brand is the product.

This is precisely the dynamic now unfolding in knowledge work, and AI is the cotton gin at the center of it. For decades, software accelerated white-collar productivity without eliminating white-collar jobs. QuickBooks digitized bookkeeping; it did not replace the bookkeeper. Excel transformed financial modeling; it did not replace the analyst. The internet enabled legal research at unprecedented speed; it did not replace the associate reviewing documents late into the night. These were incremental gains, the human remained the bottleneck. The software was a faster gin, but someone still had to feed cotton into it.

Agentic AI removes the human from that loop entirely. An AI agent can be instructed to reconcile a full set of financial records, write and test production code, review thousands of pages of legal discovery, or clean and analyze a dataset — and it will work through these tasks continuously, without fatigue, at a marginal cost approaching zero. You no longer need a floor of junior analysts or a class of first-year associates to execute the work. You need a smaller number of experienced operators and a fleet of agents.

The economic consequence is a sharp shift in the capital-to-labor ratio across knowledge-intensive industries. The cotton gin made it profitable to deploy more land and machinery relative to skilled seed-separating labor. Agentic AI makes it profitable to deploy more compute relative to entry-level knowledge workers. The labor bill shrinks; the software bill grows; and the cost of producing the output — whether a legal brief, a financial model, or a software module — falls precipitously. The product is commoditized. And there is a critical difference from the cotton gin: Whitney’s machine automated one discrete step in one industry. AI, adaptable across domains through targeted training, can in principle be applied to any structured knowledge task. This is not a gin for cotton. It is a gin that can be reconfigured for any crop.

The commoditization of execution does not destroy value — it relocates it. As with cotton and clothing, the locus of value shifts upstream: from doing to deciding, from executing to problem-framing. The lawyer who uses an AI agent to surface relevant case law in seconds is not made redundant by that speed, but is rather freed for the judgment calls that matter: the strategic framing of an argument, the read on how a judge thinks, the client relationship that no model can replicate. The software engineer who can articulate precisely what a system needs to do, and audit the output of an agent that writes it, is worth more in a world of autonomous coding, not less. What survives commoditization is the capacity to identify the right problem, navigate ambiguity, and exercise judgment where the cost of being wrong is high. The information was never the asset. The application of information was always the asset.

A reasonable counterargument is that efficiency gains expand demand rather than reduce it, and that if AI makes each associate ten times more productive, firms will simply take on ten times more work. This is the Jevons Paradox applied to labor. When steam engines became more fuel-efficient in the 19th century, coal consumption rose because cheaper energy made more applications viable. In law, where access has long been suppressed by cost, there is a real case that AI unlocks a latent market rather than shrinks the workforce. Some categories, estate planning, small business contracts, routine compliance, almost certainly have unserved demand that cheaper AI-assisted legal services would reach.

But the optimistic scenario runs into structural limits specific to the profession. Legal demand at the high end is constrained by the number of deals and disputes in the economy, not by associate bandwidth. In this context, AI won’t generate more M&A transactions or litigation. In competitive markets, the more likely outcome is that efficiency gains compress billing rates rather than inflate headcount, as clients demand and eventually receive lower fees for work agents now do in hours.

A more fundamental concern for the industry relates to the fact that the work being automated, document review, legal research, first-draft briefs, is not just output. It is the training ground through which junior associates learn and develop the judgment that makes them valuable senior lawyers a decade later. Automate the entry level, and you hollow out the pipeline that produces the partners of the future.

All this is to say that the transition to a future dominated by AI and agents will not be smooth. The historical precedent cuts both ways. Detroit at its peak employed hundreds of thousands of workers in automobile manufacturing, anchoring a middle class across an entire region. When automation and offshoring restructured the capital-labor mix in that industry, the adjustment was not orderly. Detroit’s population fell from nearly 1.9 million in 1950 to under 700,000 today.

Flint, once a crown jewel of General Motors, never recovered. The disruption posed by agentic AI is likely to unfold faster than any of those prior waves. Automotive robotics took a generation to displace manufacturing workers, but AI is being deployed at software speed, with updates that would have required years in a hardware cycle now pushed in weeks. The displacement of entry-level knowledge workers in law, finance, and software development may compress into a window far shorter than any prior labor market transition.

The future being built by AI is not unprecedented in its broad structure. Technology has always restructured the capital-labor mix, always pushed value upstream, and always rewarded those who moved with it. What is unprecedented is the scope and the speed, and the risk is that the creation of new value and the displacement of existing workers will proceed at different speeds, and across different demographics, leaving millions of jobs displaced before any retraining infrastructure exists to absorb them. The question is whether individuals, companies, and governments can adapt quickly enough to ensure the gains are broadly shared, and that the ghost towns left behind this time are fewer than the ones left before.

Sometimes you really have to wonder what life was like before the internet, when commerce meant physically going to a store, when discovery meant walking through a mall, and when information asymmetry between buyers and sellers was simply accepted as a cost of doing business. The last two decades were genuinely transformative, not because technology eliminated human behavior, but because it rerouted it.

The question on everyone’s mind now is whether AI represents another such inflection point. As models have advanced rapidly across language, reasoning, and computer vision, a debate has emerged about what AI will actually change, and more importantly, what it won’t. The sentiment has shifted noticeably in recent months, from skepticism toward something closer to fear. Everyone is fearing that agentic AI will disrupt knowledge work, compress labor markets, and dissolve the familiar structures of commerce and society.

I want to make the argument that AI will be at least as revolutionary as the internet, and perhaps more so, especially in the domain of knowledge creation and research, but this does not mean it will reduce all friction. The crucial insight that tends to get lost in this debate is this: what determines friction in any market or human system is ultimately the human being at the center of it. Technology does not decide how much friction exists. Humans do. And in some cases, AI will not reduce friction at all — it will deepen and enrich it.

The Internet’s Lesson: Technology Serves Human Preference, Not the Other Way Around

To understand where AI is headed, it helps to revisit what the internet actually disrupted. Amazon is the canonical example. Bezos started with books not because he loved literature but because books are uniform in size, making logistics predictable and cheap. But books were really just a proof of concept for a much bigger idea that you could rebuild the entire architecture of retail around a warehouse model rather than a storefront model.

Traditional retailers are constrained by physical space in expensive, high-footfall locations — a Macy’s on Fifth Avenue can only stock so much, and every square foot of floor space carries a cost. This forces a constant, high-stakes gamble of predicing what consumers will want months in advance, manufacture it, ship it to shelves, and hope you got it right. Get it wrong and you’re either sitting on unsold inventory that you have to discount or discard, or turning away demand you can’t fulfill.

Amazon’s insight was that warehouses located outside of cities, where land is cheap, fundamentally change this equation. You no longer need to guess with the same rigidity, because you’re not constrained by expensive retail floor space. You can hold more SKUs, stock up dynamically when demand signals suggest a surge is coming, and replenish far more fluidly than any traditional retailer ever could. Combined with the data flowing in from online behavior — what people are browsing, saving, and buying — you shift from a guessing game to something approaching a science. Supply begins to track demand rather than anticipate it blindly, waste shrinks, and the economics of the whole operation improve dramatically.

What the internet solved, at its core, was an information asymmetry problem. Retailers operating blind were producing deadweight loss — excess inventory, mismatched supply and demand, missed sales. Better data collection meant you could move from guesswork to something closer to science. The result was a seismic shift: fast fashion, e-commerce, and ultimately ultra-fast platforms like Shein that respond to consumer trends in near real-time. Traditional retailers like Macy’s and Nordstrom, unable to compete on price, discovery, or convenience, collapsed under the weight of their own friction.

The critical insight the age of internet revealed is that not all friction disappeared. The stock market embraced algorithmic trading and commission-free brokerage almost immediately — frictionless transactions made obvious sense for a standardized, information-dense market. Yet real estate, another information-rich market, remains stubbornly dominated by brokers, agents, and negotiation. Why? Because buying a home is an emotional act. People want a human layer of trust, guidance, and accountability precisely because the stakes are high and the decision is deeply personal. The friction isn’t a bug in that market. It’s a feature that humans have chosen to preserve.

This is the insight that gets buried in most conversations about AI. The human is the bottleneck, and that is not going away.

Agentic AI and the New Information Layer

What agentic AI promises is the internet’s core value proposition — closing information asymmetry between buyers and sellers — but at a dramatically higher resolution. Today’s e-commerce still operates with significant blind spots. A consumer browsing Abercrombie & Fitch leaves behavioral traces, but the brand still doesn’t truly know who is behind the screen, what they’re genuinely willing to pay, or whether they’re likely to respond to a complementary recommendation. The consumer, meanwhile, is limited by their own awareness — they only discover what they already know to look for.

An AI agent changes this by acting as a digital twin on behalf of the consumer. Tell your agent you want a cashmere sweater in neutral tones between $500 and $1,500, and it canvasses the entire market, surfaces brands you’ve never encountered, and returns a curated shortlist. On the seller’s side, the agent understands not just what you’ve bought before but what your preferences are now, enabling targeting that is genuinely personalized rather than statistically approximated. This exchange between agents — buyer’s agent and seller’s agent — creates a far more efficient marketplace, one where supply and demand can approach true alignment, where obscure brands can access new audiences, and where product quality competes on its own merits rather than on brand recognition or marketing spend alone.

The second-order effect of this is that opacity becomes harder to sustain. Products that rely on information asymmetry — marketing-driven premiums over actual quality, brand prestige masking generic manufacturing — become more vulnerable when consumers have agents doing rigorous comparison on their behalf. This dynamic extends beyond retail into healthcare, education, financial services, and anywhere that knowledge gaps between provider and consumer have historically generated inefficiency or exploitation.

But the Human Remains the Bottleneck

Where the standard AI narrative tends to go wrong is that it assumes that because AI can eliminate friction, it will — uniformly, inevitably, across every domain. The history of the internet tells a different story.

The stock market runs at speeds no human can perceive because humans collectively decided that speed and efficiency were the only things that mattered there. Real estate hasn’t been similarly disrupted because humans collectively decided that the emotional weight of the transaction requires friction — requires a trusted intermediary, a negotiation process, a human hand to hold through the uncertainty.

Luxury goods are perhaps the clearest example. A Hermès bag is not a product, it is an experience, a social signal, a relationship with a brand. The friction of acquiring it, the waitlist, the boutique visit, the relationship with a sales associate, that is not inefficiency to be engineered away. That is the product. An AI agent that could instantly source a comparable luxury item at optimal price would be solving a problem that luxury consumers don’t have, because the journey is the point. AI in this context doesn’t reduce friction — it could actually be deployed to enhance and curate that experience, making it richer, more personalized, more exclusive.

This suggests a more nuanced framework for thinking about where AI will and won’t be transformative. AI will eliminate friction wherever humans have decided friction is wasteful, and it will enrich and deepen friction wherever humans have decided friction is meaningful.

Routine, low-emotional-weight transactions — grocery shopping, comparing insurance plans, sourcing commodity products — will trend toward frictionless automation. The human bottleneck in these domains is already yielding, because people have little attachment to the process itself. But high-stakes, emotionally resonant decisions, like buying a home, choosing a school, navigating a health diagnosis, purchasing art or luxury goods, will retain their human texture, and AI will more likely serve as a sophisticated support layer than a replacement for human judgment and feeling.

What Kind of Friction Do We Want?

There is a risk in the frictionless ideal that doesn’t get discussed enough. A world in which your AI agent knows your preferences and curates everything to match them is also a world of narrowing, of algorithmic bubbles, reinforced tastes, and the gradual erosion of the productive discomfort that comes from encountering something unexpected. Discovery, in the truest sense, often requires friction. Stumbling into a bookstore and leaving with something you never would have searched for is a friction-dependent experience. Whether agentic AI can preserve that kind of serendipity while eliminating wasteful inefficiency is one of the genuinely open questions about the technology.

What is clear is that AI will not be a universal solvent for human friction. It will be a powerful tool that humans will use, shape, and selectively resist according to their own values, emotions, and cultural preferences. The businesses that will thrive are those that understand this distinction, that recognize where their customers want efficiency and where they want to be held, guided, and delighted by the process itself. Those who mistake the elimination of friction for a universal good will find themselves optimizing away the very things that made them worth choosing in the first place.

The human is the bottleneck. That, more than any technical capability, will determine what AI actually changes.

That feeling — that specific, quiet despair of watching homes being financially out of reach — explains why so many young people are gambling on zero-day options, piling into meme coins, splurging on $400 concert tickets and $6,000 trips to Japan, and generally behaving in ways that look, from the outside, like financial nihilism.

The numbers are almost too bleak to recite. The median age of a first-time homebuyer in the United States hit 56 years old in 2024, up from 45 in 2021 and 31 in 1981. The monthly cost of owning a median-priced American home, as a share of median income, has ballooned from roughly 30 percent a decade ago to nearly 50 percent today. Nearly half of Gen Z agrees with the statement: “No matter how hard I work, I will never be able to afford a home I really love.” Not a mansion. Not a place with a yard. Just a home they love.

This isn’t just a housing problem anymore. It’s a psychology problem, and it’s beginning to show up in the data in ways that should alarm anyone paying attention.

Bloomberg, using the Federal Reserve’s Survey of Consumer Finance data, built a model tracking financial trajectories from age twenty to seventy-five. The findings were stark: people who forgo homeownership end up with significantly less net worth over their lifetimes. That part isn’t surprising — home equity has long functioned as the primary savings vehicle for middle-class families. What was surprising was what people who gave up on homeownership did instead. Renters with net worth under $300,000 reported higher rates of cryptocurrency participation and lower overall work effort than homeowners at comparable wealth levels. The model suggested that “giving up” doesn’t just change where the money goes — it changes how people think about money, work, and the future altogether.

Behavioral economists have a name for the mental state that drives this shift. When people perceive themselves as already operating in the domain of losses, when they believe the baseline outcome is bad, they become dramatically more willing to accept risk. Not because they’re optimistic about the upside, but because the downside already feels like a given.

The money not being saved is flowing somewhere, and it’s not just gambling on crypto and Kalshi. It’s also experiences. The travel industry has documented a sharp surge in what analysts are calling “revenge spending” among younger consumers, but to those doing it, it feels less like revenge and more like a rational reallocation. If the long game is off the table, the short game becomes everything. The dinner, the trip, the concert, the moment — these things are real in a way that a 30-year mortgage increasingly is not. There is something philosophically coherent about this, even if it is also a little heartbreaking.

What makes all of this genuinely dangerous — not just personally, but socially — is what it means for the structures that hold communities together. And here it helps to think about why humans built communities in the first place.

Humans are a species that survived by cooperating. Not metaphorically, but literally — ancestors made it through ice ages and predator-heavy landscapes by hunting in groups, building shelters together, raising children collectively, and maintaining bonds across generations. That cooperation required investment in the future. It required believing that the future was worth investing in. The psychological architecture that made civilization possible, the willingness to sacrifice now for gain later, to plant crops you might not harvest, to build structures you might not live to see completed, was sustained by the belief that you had a stake in what was being built.

Homeownership, for all its complexity as a financial instrument, has historically served as a powerful proxy for that stake. When you own the land you live on, the future is not an abstraction. It’s the garden you’re planning, the school district you’re researching, the neighborhood association meeting you’re slightly annoyed about but still show up to. And when that anchor disappears, something subtle but profound shifts in the way people orient themselves toward time.

This is not an abstract concern. It is already playing out in real time, most dramatically in East Asia, where the housing affordability crisis is layered on top of brutal working cultures and economic uncertainty in ways that have produced outcomes nothing short of catastrophic.

South Korea’s fertility rate fell to 0.72 in 2023, the lowest ever recorded for any country, anywhere. Japan’s population has been declining for over a decade with no meaningful reversal in sight. China, despite its scale, is projected to lose hundreds of millions of people over the coming century, and its younger generation has coined the term “lying flat” — tang ping — to describe the deliberate choice to stop striving in a system they believe is rigged against them. These are not lazy generations. These are generations that ran the math, didn’t like the answer, and stopped playing a game they had no reasonable chance of winning.

The early signals of the same logic are visible in North America. Marriage rates among Gen Z are down. Birth rates have fallen to historic lows. Polls consistently show that a significant portion of young Americans and Canadians say they are choosing not to have children not primarily for ideological reasons, but because they cannot imagine affording a stable enough life to raise one well. The home — that physical, tangible place that has always anchored the idea of family — has become, for many, a kind of fiction.

The dream was never really just about a house. It was about the idea that effort compounds, that sacrifice means something, that the life you build is actually yours to build. Economies, as much as they run on data, also run on belief — the expectation of inflation shapes inflation; the expectation of growth shapes growth. The same is true of social contracts. When enough people quietly conclude that the contract is broken, their behavior changes in ways that make it break a little more, and then a little more, until the idea of repair starts to feel as fantastical as the listings themselves.

The housing supply crisis is not resolving any time soon. Structural shortages, a lack of skilled trades workers, restrictive zoning, decades of underbuilding, mean that meaningful relief is likely a generation away at minimum. In the meantime, a generation of people who should be planting their feet into the ground are floating, spending, gambling, checking out, and quietly mourning something they were promised and never received.

Assume three of your male friends and three female friends are single and looking for a partner. How would you arrange their first date as a matchmaker?

The best way is to have each of them rank preferences of their potential match from #1 to #3 and pair them up based on their preferences, as you ideally want to maximize the aggregate happiness of this set up.

The Gale-Shapley algorithm examines matching as a math problem, and is a matchmaking system that pairs two groups, which in our example, are men and women looking for partners. Each person ranks their preferred partners, and the algorithm finds a “stable” matching in which the aggregate happiness is maximized when no one would be better off swapping partners with someone else. What’s surprising is that who gets to make the proposals dramatically changes the outcome.

The proposing side makes offers to their top choices. The receiving side tentatively accepts the best offer they’ve gotten so far, but can always trade up if someone better comes along. Rejected proposers move down their list and try their next choice. This continues until everyone is matched. A key insight that is derived from this process is that proposers get to “shop around” starting from their dream match, while receivers can only choose among whoever approaches them.

Let’s say three women and three men have different preferences. When women receive proposals from men, each woman can only choose from the men who approach her—she might end up with her third choice simply because her preferred partners never proposed to her. But if women were the ones proposing, each might end up with her first or second choice because she could pursue her favorites directly. In both cases the matchings are equally “stable,” but the outcome favors the side that does the proposing.

This has real-world consequences. For decades, the medical residency match used this algorithm with students proposing to hospitals, systematically giving students better placements than if hospitals had proposed. The algorithm isn’t neutral; it’s a powerful tool that favors whoever holds the right to propose, even while appearing fair on the surface.

The Gale-Shapley algorithm succeeds in matching medical residents to hospitals and students to schools because it operates under crucial constraints that modern dating fundamentally violates. In functional matching markets, participants face limited capacity, forced commitment, and bounded choice sets. These constraints force realistic preference formation and eventual acceptance of good matches.

Modern dating, as evident with online dating, creates an illusion of unlimited options where users can perpetually swipe through thousands of profiles, top-tier individuals face no capacity constraints forcing them to commit, and no mechanism exists to bind matches into actual relationships. The algorithm assumes people will eventually accept their best available match, but dating apps allow, and profit from, users rejecting viable matches indefinitely while chasing unrealistic ideals.

This dynamic creates the “abundance mindset paradox.” Before modern dating, geographic and social constraints naturally created small, bounded pools of potential partners, perhaps 50-200 people you’d realistically meet through work, school, or social circles. Within these limited pools, people naturally calibrated their expectations through feedback and social proof, learning their approximate “market position” and accepting good matches accordingly.

Online dating and social media together destroyed these natural constraints by providing complete market information. Users can now see profiles of tens of thousands of potential partners on dating apps while simultaneously being exposed to carefully curated highlight reels of the most attractive and accomplished people on Instagram, TikTok, LinkedIn, and other platforms. Dating apps show you the top 1% of potential partners in your area, while social media shows you the top 0.01% of attractive people globally, creating a double layer of unrealistic benchmarking.

This constant exposure fundamentally warps preference formation. A person who might happily date someone in the 60th percentile of their local pool now compares them not just to the 90th percentile on dating apps, but to the algorithmically-curated “perfect” bodies and lifestyles flooding their social media feeds. The result is rejection of realistic local options while pursuing dating app matches in the top tier—who themselves are chasing even higher, influenced by the same social media exposure—and who, drowning in options and comparing their matches to Instagram ideals, never commit to anyone. Social media doesn’t just show you better options, it manufactures an entirely fictional standard of what you “deserve” based on people whose attractiveness, wealth, and lifestyle are often professionally produced, filtered, and completely unattainable for 99.9% of the population.

This exposure calcifies into rigid, non-negotiable requirements that further break the matching mechanism: the person must be over 6 feet tall (eliminating 85% of men), earn over $100,000 annually (eliminating 90% of people), and have model-level looks (eliminating 99%). These fixed parameters, often targeting characteristics that describe less than 1% of the population, aren’t presented as preferences but as minimum standards, below which a person is considered “settling.”

The cruel irony is that being realistic about partner selection has become culturally conflated with choosing a subpar partner, when in reality it’s simply acknowledging market constraints. Just because you’re a straight-A student doesn’t mean you automatically get into Harvard; you might end up at UC Berkeley, an excellent school, simply because so many qualified students are competing for limited Harvard spots.

This isn’t to say you should settle for a community college if you’re a genuinely brilliant student, as the Gale-Shapley algorithm is designed to get you the best match you can realistically attain, not to force you into an incompatible pairing. The point is recognizing that UC Berkeley, while not Harvard, is still an outstanding institution where you can thrive, just as a partner in your realistic range, say, the 60th-80th percentile if you’re in the 70th, can be genuinely wonderful, even if they’re not the Instagram-perfect ideal.

The quality of both education and relationship depends far more on what you make of it than the brand name or superficial metrics. A partner who is 5’10”, earns $75,000, and is genuinely kind and compatible might create a far better relationship than the 6’2” high-earner you’re endlessly pursuingm, but you’ll never discover this if you filter them out entirely based on height and income requirements derived from social media fantasies.

The equivalent of refusing to attend any university because you didn’t get into Harvard is precisely what’s happening in modern dating. People are choosing to remain unmatched indefinitely rather than “settle” for partners who don’t meet arbitrary ideal standards and never realizing that the ideal itself is a statistical impossibility and that fulfilling relationships.

Viewing the ideal as the realistic discounts what matters most in a relationship—finding a partner whose values align with yours, who is supportive and honest, who pushes you and challenges you to grow together, and who accepts you at your best and your worst.

As Shakespeare wrote in Sonnet 116:

Let me not to the marriage of true minds

Admit impediments; love is not love

Which alters when it alteration finds,

Or bends with the remover to remove.

O no, it is an ever-fixèd mark

That looks on tempests and is never shaken;

It is the star to every wand’ring bark

Whose worth’s unknown, although his height be taken.

Love’s not time’s fool, though rosy lips and cheeks

Within his bending sickle’s compass come.

Love alters not with his brief hours and weeks,

But bears it out even to the edge of doom:

If this be error and upon me proved,

I never writ, nor no man ever loved.

Shakespeare understood four centuries ago what matching theory proves today—that true worth cannot be measured by superficial characteristics we use to filter partners. The algorithms that work succeed because they prioritize compatibility and stability over the endless pursuit of an unattainable ideal. The best match isn’t the one that looks perfect on paper, but the one that proves steadfast “even to the edge of doom.”

But beneath the surface of this aesthetic perfection lies a darker truth, one that became impossible to ignore in the winter of 2025.

In February 2025, Japan’s Ministry of Health, Labor, and Welfare released statistics that came as a surprise to nobody. The number of births in 2024 had fallen to 686,061, the first time in recorded history that the figure dropped below 700,000. It marked the ninth consecutive year of decline, accelerating 15 years ahead of expert predictions. The natural population decline hit a record 919,237. Deaths climbed to 1,605,298. During the first baby boom following World War II, Japan saw 2.5 million births annually. Now, fewer than 700,000 new lives arrive while the grave claims over 1.6 million. The total fertility rate fell to 1.15. In Tokyo, it dipped to 0.96.

These aren’t just statistics. They’re the death rattle of a society that has stopped believing in the future.

The timing of these demographic revelations coincided with another crisis. In October 2025, Japan elected its first female prime minister, Sanae Takaichi, a hardline conservative protégé of the late Shinzo Abe. Her historic ascension should have been cause for celebration. Instead, it triggered a market convulsion. Takaichi had campaigned on “responsible and proactive” fiscal expansion—a proposed two-year suspension of the 8% consumption tax on food, higher defense spending, and aggressive industrial policy. The market’s response was brutal. Japan’s 10-year government bond yield pierced 4% for the first time in decades.

The skepticism was warranted. Japan’s government debt stood at approximately 235% of GDP as of March 2025. The Bank of Japan alone held 46.3% of government debt. Japan’s economy had finally begun normalizing after decades of deflation, and the central bank was poised to raise rates. Takaichi’s expansionary fiscal policy and the bank’s tightening monetary policy were on a collision course.

But the real question no one wanted to ask was this: Could fiscal stimulus resurrect a society that had already decided to die?

To understand Japan’s predicament, you have to travel back to 1989, the year the bubble burst. Between 1989 and 1992, the Nikkei 225 plummeted over 60%. Urban land prices collapsed by 15-20%. The country entered the “Lost Decades,” a prolonged stagnation from which it never truly recovered. In response, policymakers turned to quantitative easing, turbocharged under Abe into “Abenomics” in 2013. The policy boosted stock markets, but Japan’s debt burden soared past 200% of GDP and kept climbing. The fundamental issue: monetary and fiscal interventions were treating symptoms rather than causes rooted in culture and institutions.

I visited Japan briefly last winter, in Tokyo. What struck me most was the overwhelming sense of order and hierarchy. The conformity was immediately visible on the subway—everyone dressed in neutral colors, the styling remarkably uniform. I thought about the New York City subway, a kaleidoscope of chaos. In Tokyo, you could feel the weight of social expectation pressing down.

One evening at midnight, I found myself eating katsu don at a small restaurant. Two men in business suits sat nearby, hunched over laptops, still grinding away at work. Locals confirmed that the culture of endless work hours still permeates Japanese society. But the most jarring discovery was the women-only train cars—existing because groping and harassment on crowded commuter trains is persistent enough to warrant segregated transportation. While some celebrate this as advancing women’s safety, it struck me as a symptom of a deeper pathology.

A Japanese friend once explained this paradox: “Japan only evolves during times of crisis. When the threat of imminent danger fades, we stagnate.” The country functions like a single-minded organism. As anthropologist Ruth Benedict wrote: “The Japanese, more than any other sovereign nation, have been conditioned to a world where the smallest details of conduct are mapped and status is assigned...they learned to identify this meticulous plotted hierarchy with safety and security.”

But what happens when that world no longer offers a future worth participating in?

In macroeconomics, the household is the basic unit of measurement. It’s where consumption happens, where future workers are born and raised. The formation of households builds social capital that translates directly into economic capital. So why have Japanese people stopped forming households?

Japan’s working-age population peaked in 1995 and has been declining ever since. Despite government policies aimed at encouraging childbearing—subsidies, expanded childcare programs, aggressive matchmaking support—births continue to fall. While marriages edged up 2.2% to 499,999 in 2024, this came only after steep declines. And unlike in Western countries, only a few out of every 100 babies in Japan are born out of wedlock.

The reality is that many Japanese people, particularly young people, are engaging in passive resistance. They’re opting out through the quiet refusal to participate in the traditional life script: career ladder, marriage, children, retirement. This isn’t laziness. It’s a rational response to a system that demands total conformity but offers diminishing returns. Long work hours, rigid hierarchies, limited career mobility for women, social stigma around alternative lifestyles, astronomical costs of living—all compound into a simple calculation: Why bother?

When a society makes the basic act of family formation so difficult, so financially precarious, so socially constrained that large swaths of the population simply opt out, you’re witnessing the slow collapse of collective will.

Enter Takaichi in October 2025, promising a revival through fiscal firepower. But Abenomics already ran this experiment for nearly a decade. It boosted stock prices, weakened the yen, kept the government afloat. What it didn’t do was address the fundamental problem of a society whose institutions and cultural values are frozen in amber, unable to adapt to new realities.

The most apt comparison is to a patient with chronic illness who doesn’t believe treatment will work. Without the patient’s active participation in recovery, nothing changes. Takaichi’s fiscal expansion will likely follow Abe’s trajectory, led by temporary market reactions, mounting debt, and no fundamental shift in demographic trends. You cannot spend your way out of a crisis of meaning and purpose.

Economics is rooted in culture and institutions. When those cultural values and institutional structures calcify rather than adapt, economic policy becomes like painting over rust. It might look better temporarily, but the underlying decay continues. Japan’s future, absent fundamental transformation, is one of continued population decline, mounting debt, and a society that grows grayer in every sense. Not through catastrophe, but through the quiet withdrawal of a population that has collectively decided there’s no future worth investing in.

Some observers call this “late-stage capitalist decline.” I think it’s what decline looks like in general. Decline is when a society becomes unable to evolve, when it constrains its ability to voice dissent and innovate. Decline is when social norms are challenged but unsuccessful. Decline is passive resistance writ large—the decision to stop playing the game entirely.

Japan’s tragedy is particularly poignant because the country has so much to offer. The craftsmanship, the aesthetic sensibility, the emphasis on quality and care. But they’ve become trapped in a social structure that punishes deviation and rewards conformity to the point of self-destruction.

The country is becoming what I saw on that subway. A sea of black and gray, individual color drained away; everyone moving in the same direction toward a destination nobody really wants to reach.

The AI lansdcape has shifted. The industry is no longer talking about whether AI delivers value, but rather grappling with how fast it’s reshaping entire industries. The implication of this shift is that the market doesn’t know how to price that uncertainty.

While consumer AI fumbled its way through 2023, a different story was unfolding in enterprise. Anthropic emerged not just as a competitor, but as the leader for serious business applications. With roots in applying LLMs to coding and an unwavering emphasis on safety and compliance, Anthropic became the natural choice for companies navigating regulatory requirements and sensitive data.

AI models have become tools extending beyond coding into entry-level knowledge work across finance, law, and business operations. Gamma reimagining PowerPoint. Rogo acting as your investment banker. Harvey as your legal assistant. Each startup carved out a niche, applying AI in refined, articulate ways to solve real problems.

This was AI’s original promise: a productivity catalyst. And for the first time in decades, the numbers proved it. US GDP broke free from its historical trend. Productivity levels surged. Companies stopped hiring and started cutting, particularly at entry levels.

The Shopify CEO’s internal memo crystallized the shift perfectly: “Employees will be expected to prove why they ‘cannot get what they want done using AI’ before asking for more headcount and resources.”

Then Anthropic accelerated this trend with its latest releases. Cowork. Plugin for Excel. Legal industry solutions. Each release wasn’t just an incremental improvement, but rather a fundamental challenge to how work gets done.

The market’s reaction? Panic.

This is the Anthropic moment, the realization that traditional software economics are breaking down. Software has always been seat-based. What happens when software can be built cheap and fast? Do companies still need as many seats? Do they need to outsource software development at all?

Unlike the fleeting DeepSeek moment, the Anthropic moment isn’t about a single breakthrough or model release. It’s about the cumulative impact of sustained improvements that have reached an inflection point, where adoption into enterprise workflows and knowledge work is accelerating at an impossible-to-predict pace.

The economic logic is brutal. Would you hire a new grad for $100k annually to do data analysis when you can equip existing employees with Claude to run agents and complete the same tasks? Scale that decision across an organization, then across entire industries. You’re looking at trillions in cost savings and productivity gains.

So companies are investing. Cloud providers are pouring in billions to meet demand. Google Cloud accelerated to 48% growth from the high 30s, capital expenditure jumping from $119 billion to $180 billion. Amazon pushed capex above $200 billion. Together, these two companies alone are investing roughly 1% of the entire US GDP into AI infrastructure.

Yet the market sold off.

Google dipped after earnings despite fantastic cloud growth. Amazon dropped over 10% post-earnings despite AWS acceleration. Even Nvidia traded down in after-hours, despite being the direct beneficiary of hyperscaler capex flowing into computers.

The rotation wasn’t into other tech. It was into defensives—consumer staples like Walmart. Industries AI won’t disrupt.

I can only find a single rational justification to this selloff: the Anthropic moment has injected massive uncertainty into the tech narrative.

After three years of strong returns, investors simply don’t know what the landscape will look like in 18 months. How quickly can these models scale? How fast will enterprises adopt them? Which entire software categories will become obsolete? The shock in Alphabet’s earnings commentary said it all.

Markets need certainty to justify long-term investment decisions. Without it, risk demands a higher premium and stocks get re-rated lower.

The semiconductor sell-off reflects profit-taking and hedging against potential software sector losses. The hyperscaler sell-off stems from capex concerns in an environment where no one truly understands the payback timeline or competitive dynamics.

What we’re witnessing isn’t a bubble bursting. It’s the opposite. AI is finally delivering on its promises so effectively that the market can’t model the disruption.

The Anthropic moment is the point where improvements in AI capabilities have accumulated to reach a tipping point that widespread enterprise adoption is no longer a question of “if” but “how fast.” It’s where the technology moves from experimental to existential for traditional business models, espeically in industries vulnerable to AI disruption.

And unlike the DeepSeek moment that captured attention briefly before fading, the Anthropic moment is structural. It’s not going away. It’s accelerating.

The market’s confusion is rational. When you can’t predict which industries will be transformed, which companies will adapt, and which will become obsolete, all within a compressed timeframe, the prudent move is to de-risk.

But make no mistake, the Anthropic moment is here. The technology has crossed the threshold from promising to proven. The productivity gains are real. The disruption is underway.

And this time, it’s here to stay.

Would you have still bought it? Hell no. Unless you’re already planning your funeral, you’re demanding your money back or a massive discount. Because the game just changed overnight.

Now replace “Vancouver home” with “software company stocks” and “earthquake” with “AI.” Congratulations, you’re a Wall Street investor in February 2026. And you’re absolutely panicking.

On Tuesday, February 3rd, $285 billion evaporated. Gone. In one day. Thomson Reuters down 16%. Legalzoom down 20%. The software index posted its worst day since October 2008, a 15% nosedive.

And what caused this carnage? A blog post.

Anthropic, the company behind Claude AI, quietly released a tool that automates legal work. Just dropped it on their website. No fanfare. No press conference. Before the opening bell even rang, traders were hitting the sell button like it was on fire. The panic wasn’t about one legal tool. It was the sudden, visceral realization: “Oh shit, if AI can do this to lawyers, nobody’s safe.”

The bloodbath spread faster than wildfire. Software loans collapsed. Private equity titans who’d been gorging on software deals at absurd valuations got destroyed. Blue Owl Capital, which basically built its empire betting on software, crashed 13% to a nine-day losing streak record. Ares, KKR, TPG? All down 10%+. These are firms managing trillions. And they’re getting absolutely wrecked.

Software represents 12% of the entire leveraged loan market. That’s not a niche corner. That’s the foundation. And it’s cracking. UBS analysts are now estimating default rates could hit 13%. To put that in perspective, that’s nearly triple the normal rate. Banks are quietly freaking out.

Now, the defenders will tell you everyone’s overreacting. The Economist literally made fun of the panic, pointing out that even OpenAI uses Slack (owned by Salesforce). Jensen Huang from Nvidia called it “exaggerated,” insisting AI turbocharges software rather than destroys it.

Cool story. Except the market doesn’t care about your hot takes. The market cares about numbers. And the numbers are brutal.

This quarter, only 67% of software companies beat revenue expectations versus 83% for tech overall. Microsoft reported decent earnings and still dropped 10% in a day because cloud growth is slowing. January was its worst month in over a decade. SAP mentioned “slight deceleration” and fell 15%. ServiceNow beat estimates and still dropped 13%.

These aren’t startups burning through venture capital. These are the giants. The “indispensable” ones. Adobe for creatives. Workday for HR. Salesforce for literally everything. If you work at a company, you’re probably using their software right now.

But here’s the thing about “indispensable”—that’s exactly what everyone said about IBM 50 years ago. IBM owned mainframes like Apple owns iPhones. Suggesting IBM wouldn’t be the world’s most valuable company would’ve gotten you laughed out of every boardroom in America.

Then personal computers happened. Then the internet. Then smartphones. Each wave brought entirely new winners: Dell, Microsoft, Apple. IBM? Still around. Still making money. But completely irrelevant compared to what it was. A shadow. A reminder that dominance is temporary when the technology shifts.

That’s what keeps investors up at night. Not that Salesforce disappears tomorrow. But that AI represents a fundamental shift, a new game where all the walls the incumbents built over decades become meaningless. Where startups can simply go around them, like how the Germans bypassed the Maginot Line that the French built after World War I by going through Belgium.

And one company is already proving this thesis works: Palantir just reported 70% revenue growth and guided for 61% next year. Seventy. Percent. Meanwhile, where’s Salesforce’s acceleration? They’ve been hyping their AI agent “Agentforce” for a year. Where’s their explosive growth?

Crickets.

That’s why the market is dumping these stocks. Not because of irrational fear. Because these companies trade at premium valuations as growth stocks, keep promising AI will save them, and then... nothing. No acceleration. No proof. Just vibes and PowerPoints.

Meanwhile, Anthropic’s Claude Cowork tool, released in January, lets people with zero coding experience build apps. The barrier to entry just collapsed. Morgan Stanley, an investment bank, literally told clients to buy junk bonds instead of software company loans. Think about that recommendation. JUNK BONDS are safer than lending to software companies. That’s not panic. That’s rational fear.

A trader at Jefferies summed it up perfectly: “People are just selling everything and don’t care about the price.” When traders stop caring about price, you know things are bad.

But what’s interesting is that some companies are getting unfairly destroyed. Duolingo built its new chess product in under 11 months using AI. The company is actually using AI to innovate faster. But it’s getting lumped into the “software is dead” narrative anyway. The market is throwing babies out with bathwater, as one analyst put it.

The difference is speed. This isn’t happening over decades like the PC revolution. OpenAI is burning $17 billion this year alone. The scale of investment is orders of magnitude larger than anything during the dot-com boom. This meteor isn’t approaching—it has already hit.

Software stocks had their worst January since 2008. Worse than COVID. Worse than 2022’s tech selloff. The sector is more oversold than at any time since 2018. Some fund managers are buying, betting on a bounce. Even Microsoft is trading at its cheapest valuation in three years. But even the bulls admit recovery will take years.

Meanwhile, Blue Owl Technology Income Corp.—a fund focused on software companies—let investors pull out 17% of assets, triple the normal limit. Redemptions hit 15.4%. Translation: people are running for the exits.

One restructuring advisor nailed it: “The businesses aren’t broken. The balance sheets are just too stressed.” Companies borrowed billions at peak valuations assuming perpetual growth. Now growth is slowing, interest rates are higher, and those debt payments are crushing them.

This is what happens when earthquake risk gets repriced overnight. Except we’re not talking about a few houses in Vancouver. We’re talking about a multi-trillion-dollar industry getting fundamentally revalued in real-time.

What we’re witnessing isn’t a bubble. It’s proof that AI is real and moving faster than anyone expected. The promise of replacing knowledge work is happening now. Some software companies will adapt and survive. Many won’t. And right now, the market can’t tell which is which.

So it’s selling everything. Asking questions later. Recalculating who makes it to 100 and who doesn’t survive the next two years.

$285 billion gone in a day isn’t panic. It’s the market doing math. Cold, brutal, unforgiving math.

The earthquake already happened. Now we’re just counting the casualties.

Welcome to the SaaSpocalypse.

Want to know a reliable method for appearing intellectually sophisticated in nearly any social setting? Here is the formula: simply articulate a critique of Donald Trump’s policies, character, or conduct. It has become something of a cultural reflex, criticizing the current president is widely interpreted as a mark of thoughtfulness, regardless of the substance or originality of the criticism itself.

At a recent conference, the editor-in-chief of The Economist Zanny Beddoes spent considerable time grilling JPMorgan Chase CEO Jamie Dimon on the perils of Trump’s policies. She asked directly whether the Trump administration is making the world safer and stronger, and making NATO stronger.

Jamie responded: “I do not think it is a binary thing.”

As Jamie continued explaining how difficult it is to assess policy impacts until years later, Zanny pressed harder, characterizing Trump’s approach as “a transactional foreign policy; a foreign policy that places much less weight on alliances; a bullying foreign policy.”

Bullying whom? American allies in the European Union, partners in Asia, Canada presumably. But is America really bullying, or is it desperately watching the world order it built after World War II crumble and taking a last-ditch effort to save it?

The world order as we know it emerged in the wake of World War II. Almost immediately, the world descended into the Cold War, a battle for supremacy between the United States and the Soviet Union. While the war appeared to be a competition for military dominance, what the two superpowers were really battling over was which economic system would lead the global order.

The Soviet Union was what Mao would call a “paper tiger”—something that appears powerful but is actually ineffectual. By 1989, the United States GDP was approximately 5.6 trillion dollars compared to the Soviet Union’s estimated 2.5 trillion dollars, more than twice as large. The United States heavily fortified Western Europe through the Marshall Plan, which provided over 13 billion dollars to rebuild European economies between 1948 and 1952. The Soviet Union disintegrated in 1991, cementing the preeminence of the American century.

While much of the world’s focus in the 2000s was placed on the United States spreading its influence in the Middle East, a more important development was happening at home: the internet and personal computing revolution. You saw the birth of generational companies during that period: Apple, Amazon, Alphabet, Microsoft, PayPal, Facebook, and Nvidia—now the leader in powering today’s artificial intelligence revolution. While American involvement abroad showed the limits of military intervention, in the digital realm that frontier was infinite.

This pioneering lead made the United States a superpower. Beginning in the 2000s, the United States equity market has averaged a cumulative annual return of approximately 10 to 12 percent compared to around 5 percent for global markets excluding the United States, with performance heavily dominated by technology. This revolution is happening again today with artificial intelligence, as United States GDP has been revised upward. Recent quarters have shown GDP growth around 3 to 4 percent, driven by strong artificial intelligence investment and productivity gains. Elsewhere in the world, economies remain subdued compared to the dynamism and resilience of the American economy.

And that is a problem.

For the free world to thrive, you must have vibrant economies and trade. But the European Union is weakening. As Canadian Prime Minister Mark Carney called it, “we are in the midst of a rupture, not a transition.” Under this so-called rupture, the United States appears to be the villain because it is threatening tariffs and attempting to rebalance trade through economic pressure.

The question to ask is: who is not living up to their end of the bargain?

Carney himself acknowledged that under the liberal international order, the United States “provides public goods—open sea lanes, a stable financial system, collective security, and stable frameworks for resolving disputes.” But all of this is not free and comes at a cost funded with record debt spending.

The United States provided over 122 billion dollars in military, financial, and humanitarian aid to Ukraine through 2025, more than the European Union’s combined contribution. Why is not everyone else chipping in proportionally?

To be fair, many European nations are meeting their NATO commitment of spending 2 percent of GDP on defense. Some, like Poland and the Baltic states, even exceed it significantly. The problem is not that they are shirking their responsibilities by the agreed-upon metrics—it is that the total sum remains insufficient, and more critically, it is not growing at the pace required to match the threats of our time. When your baseline is small and stagnant, meeting a percentage threshold means little in a world where adversaries are expanding their capabilities exponentially.

But the deeper issue is not just about defense spending. Why is not more money being invested in developing domestic industries and technology? Some might argue that America has an advantage because it is a larger, more unified market. This argument crumbles under scrutiny. The European Union represents a market of over 440 million people with a combined GDP comparable to the United States. Size is not the issue. The issue is that European institutions are actively hostile to technology development. Rather than fostering innovation, they strangle it with regulation, bureaucracy, and risk aversion.

Consider the General Data Protection Regulation, the Digital Markets Act, the Digital Services Act, the Artificial Intelligence Act—layer upon layer of regulatory frameworks that create an environment where building the next Google, Amazon, or Nvidia is virtually impossible. European venture capital is a fraction of what flows in the United States, not because Europe lacks capital but because the institutional environment penalizes risk-taking. Start a tech company in San Francisco, and you have access to investors who understand that a few hundred failures might fund one world-changing bet. Try the same in Brussels or Paris, and you will spend more time navigating compliance frameworks than building your product.

In Canada, the story is similar. The country is dependent on financial services and natural resources. But when it comes to the tech sector, Canada has little to boast about. Instead, Canadians poured excess capital into real estate, creating a nationwide housing affordability crisis. When prices stopped rising, real estate became a net drag on growth. Really? Is this what America subsidized Canadian defense for?

As Jamie Dimon noted, America is less stable now than it used to be. It is less stable because its debt is ballooning—exceeding 36 trillion dollars in 2025—and it cannot sustain the same burden knowing that other Western nations, while meeting their formal commitments, lack the economic dynamism to shoulder a proportional share of the free world’s future. Meanwhile, China is eating their lunch as European industries face increasing threats from Chinese innovation and manufacturing prowess. A weaker Europe is a fragmented Europe, and a fragmented Europe is historically a warzone.

Yet somehow, America is cast as the villain, and Trump is labeled insane. This is not to say America has not benefited from these relationships—it has—but America cannot be the sole bearer carrying the free world on its shoulders. The unipolar moment has ended.

I do not always agree with Trump’s method of leveraging tariffs and issuing threats to compel other Western allies to make needed reforms. But let us not forget that America is not the one needing this change, as it always has an ocean to retreat behind, as it did after World War I. What emerged then was World War II. This time, the United States decided to stay engaged, to stabilize Western Europe and reject isolationism, creating the greatest prosperity ever witnessed in human history.

We stand now at one of those rare inflection points in history. The institutions that kept the peace for eighty years are groaning under the weight of outdated assumptions. The bargains struck in the ashes of World War II no longer reflect the distribution of power or the nature of threats we face. We are between two worlds.

And this is precisely where Trump is right. Not in his methods, perhaps, but in his diagnosis: the free world is in decay, and it is not because of America. The survival of the free world cannot rest on American strength alone. It requires a collective commitment—not merely to meeting technical thresholds of defense spending, but to fostering the innovation, economic dynamism, and shared sacrifice that security in the modern age demands.

Freedom is not a gift that, once given, remains forever. It is a condition that must be actively maintained, defended, and renewed by each generation. The free world only remains free when all who benefit from its order contribute to its preservation. Security cannot be outsourced. Prosperity cannot be assumed. Innovation cannot be regulated into existence.

America will survive the unraveling of the current order. The rest of the free world may not. Trump’s rhetoric may be crude, his tactics blunt, but the underlying truth remains: either the West rediscovers its capacity for renewal and shared responsibility, or the ocean that once protected America will isolate it. The free world was never a birthright. It was an achievement—one that must be earned again and again, or it will simply cease to be.

What is the fate of a middle power? Think of it like a high school lunch period, where there are a few kids who naturally gravitate to the center of the cafeteria, and everyone else figures out where they fit around them. It is not necessarily about dominance—more about gravitational pull.

Prior to this year’s World Economic Forum (WEF) in Davos, Switzerland, the conventional wisdom was depressingly simple: middle powers must accommodate the interests of great powers. It is not a particularly inspiring rallying cry, but it has been the operating manual for most of human history.

The clearest and most tragic example is the ongoing war in Ukraine, a nation that by every metric qualifies as a middle power. Before February 2022, Ukraine produced approximately 10% of the world’s wheat exports, ranked as the world’s 13th largest steel producer, and boasted a GDP of $200 billion with 44 million citizens. That prosperity, however, came with invisible strings attached to Moscow. When President Zelensky tilted toward European integration, essentially trying to switch lunch tables, Putin responded with a full-scale invasion.

If not for Russia’s shocking military incompetence and $113 billion in U.S. aid through 2024, this war would have concluded long ago. Instead, the conflict has dragged on for nearly four years, with the UN estimating over 500,000 combined military casualties and millions displaced. It is the grim reality that we scroll past while checking our phones.

But isolating Russia would be unfair given recent developments in Venezuela. In January 2025, the Trump administration facilitated the capture of Nicolás Maduro, who maintained close ties with both China (Venezuela’s largest creditor) and Russia. While this does not constitute the full-scale invasion evident in Ukraine, the message rings clear that middle powers enjoy freedom only as long as they accommodate the nearest great power.

This past week, Canadian Prime Minister Mark Carney delivered what can only be described as a magnificently passive-aggressive speech at Davos. Without naming names (though everyone knew exactly which oversized neighbor he meant), Carney articulated what middle powers whisper privately but rarely declare publicly: “If we are not at the table, we are on the menu.” His prescription? Middle powers should stop waiting for someone else to restore order and instead focus on building strong domestic economies through trade diversification with fellow middle powers rather than complete dependence on great power patrons.

The speech received a standing ovation, which is rare at Davos. Carney’s message felt refreshingly honest in a world increasingly shaped by major powers. Whether it is American influence in the Western Hemisphere, Russian designs on Eastern Europe, or Chinese assertiveness in the South China Sea, the pattern holds.

As much as I admire Carney’s vision for middle power cooperation, the world is sliding toward realpolitik, where power disparities shape outcomes. We are witnessing proxy conflicts from Gaza to Sudan. Economic warfare has become normalized, with the U.S. imposing tariffs reaching 25% on steel and aluminum imports and China retaliating with rare earth export restrictions.

The architects of the post-World War II order imagined that institutions like the United Nations (UN), World Bank, and WTO would manage competition and protect smaller states. Results have been mixed. The UN Security Council has been paralyzed by vetoes—Russia alone has exercised its veto 143 times since 1946. The WTO’s dispute resolution mechanism collapsed in 2019 when the United States blocked appellate body appointments.

One post-war achievement remains: the European Union (EU), nurtured by the Marshall Plan, through which the U.S. poured approximately $13 billion (equivalent to roughly $173 billion today) into European reconstruction between 1948 and 1952. This created interdependence that made war economically irrational. The result has been the “Long Peace”—no great power conflict since 1945, unprecedented when compared to the World Wars that killed approximately 80 million people combined.

Yet the EU also exposes the limitations of middle power cooperation. The 2008 financial crisis revealed deep fractures over Greece’s €323 billion debt. More recently, unity cracked over Russian energy dependence (Germany imported 55% of its gas from Russia pre-invasion), sanctions implementation, and immigration policy (Hungary accepted 1,294 asylum seekers in 2022 while Germany accepted 244,132).

The EU’s struggles remind us of what happens when a stabilizing force weakens. For seven decades, the United States has provided security, maintained freedom of navigation protecting $14 trillion in annual maritime trade, and offered a stable reserve currency. When that central role recedes, underlying tensions resurface.

The unsentimental truth is that the world absolutely needs middle powers for trade, resources, and innovation. Global trade flows of $32 trillion annually depend heavily on middle power participation. But we should maintain clarity about structural realities.

Perhaps the sternest reminder came from how quickly Donald Trump reversed course on Canada’s recent moves with China. In early January, Canada secured an agreement allowing resumed canola oil exports (worth potentially $2 billion annually) while agreeing to import 49,000 Chinese electric vehicles. Initially, Trump praised the pragmatism: “When you get a deal with China, you take it.” But within 72 hours of Carney’s Davos speech, Trump threatened 100% tariffs on Canadian goods with Chinese content.

Here is where we need to take a detour to talk about ferns and trees, a metaphor that captures the structural reality of middle power existence. In a forest, a fern growing under a massive oak does not compete for the same ecological niche—that is a losing game. The oak represents the great power, naturally dominating the canopy and claiming most of the available light. The fern, like a middle power, cannot challenge this fundamental arrangement. Instead, it finds scattered patches of sunlight filtering through the canopy and builds a survival strategy around those opportunities.

The oak is not actively suppressing the fern out of malice—it simply occupies the space that great powers occupy, taking what they need to thrive. The fern survives not by growing upward to compete directly, but by adapting to the shade, finding the gaps, and making the most of limited resources. This is essentially what Carney proposes for middle powers: pragmatic adaptation to existing conditions rather than futile attempts to reshape the forest itself.

Canada’s current trade position is like a fern that has only grown in one direction, leaving it vulnerable if conditions in that single patch of sunlight suddenly change. Canada’s top five non-U.S. export partners combined accounted for just $105 billion compared to $435 billion to the United States alone. Roughly 75% of Canadian exports go to a single customer, a concentration that would terrify any economics professor discussing supply chain risk.

The economic case for diversification is not about political autonomy—it is basic risk management. South Korea expanded its trade partnerships across Southeast Asia and Europe between 2010 and 2023, growing ASEAN trade from $90 billion to $191 billion and EU trade from $79 billion to $142 billion. The result? Reduced vulnerability to any single market downturn while maintaining GDP growth averaging 2.7% annually.

For Canada, the opportunities are tangible: India’s $3.7 trillion economy demands Canadian potash and pulses (lentil exports already exceed $1.7 billion annually); Indonesia’s 277 million consumers and 5.3% growth rate need agricultural products and mining equipment; and Japan’s $11.9 billion in Canadian imports could expand significantly in LNG as they transition away from other suppliers. Meanwhile, Canada’s EU trade reached $119 billion in 2023 under CETA but represents just 1.5% of total EU external trade—substantial room for growth exists as Germany seeks alternatives to Russian energy and France increases plant-based protein imports.

This is where successful middle powers like Australia offer instructive examples. Australia maintains China as its largest trading partner ($202 billion in two-way trade in 2023) despite significant political tensions, while simultaneously strengthening ties with India ($50 billion), Indonesia, and ASEAN nations collectively. This resembles a fern extending fronds in multiple directions to catch scattered sunlight filtering through the canopy, rather than trying to grow straight upward into the oak’s space.

Canada will always exist in proximity to the United States—this is simple geography and economic reality. Carney understands this, which is why his Davos speech never suggested turning away from the primary relationship. Instead, he is advocating for “constrained pragmatism,” working within natural limitations while cautiously exploring their edges.

This pragmatism will serve Canadians well in the medium term, but make no mistake—it will not fundamentally reshape a world order where great powers remain key players. The cafeteria will still have its center tables. The forest will still have its towering oaks determining how much light reaches the understory. Middle powers can survive, even occasionally thrive, by being smart, adaptable, and realistic about their constraints. What they cannot do is rewrite fundamental structural realities.

The fate of the middle power, then, is neither complete dependence nor genuine autonomy, but rather the careful work of navigating between these poles, seeking marginal improvements, building modest coalitions, and finding opportunities in the spaces that remain available.

Please note that the opinions expressed in this piece are my own and do not reflect those of my employers. Thanks for reading The Americano—don’t forget to share it with a friend!

I hate pick me girls.

“Girls are so catty and that’s why I prefer hanging out with guys.”

“I could never be into that ‘girly’ stuff like makeup and clothes. I’m just not really like other girls.”

I’ve always wondered if there could actually be a type of person in the world that throws their own kind under the bus just for attention. Like, how stupid do you have to be? We all know betrayal exists—in Shakespeare’s Macbeth, the titular character commits treason by killing Duncan to become king. But I feel like that’s somewhat justified given we are, after all, talking about becoming the king. We’re talking about absolute power. At least Macbeth was reaching for a crown, not a pat on the head.

But imagine degrading your own sex to gain the attention of men and validation. That’s what defines a pick me girl. What kind of attachment issues do you have to make that your whole personality?

Here’s the thing though, while “pick me” is typically associated with women, men can be pick mes too. And these past few months, the President of the United States has really decided to throw America under the bus (again) because he wanted attention from the world. Trump has decided that after making headlines by capturing Maduro in early January that the show must go on, with a new season and a new plot: CAPTURE GREENLAND.

I don't have to tell you how ridiculous the idea really is because Mark Carney already gave Trump a lecture in his speech delivered at the World Economic Forum. Carney's central argument cut to the bone: "you cannot live within the lie of mutual benefit through integration, when integration becomes the source of your subordination."

In other words, America has spent decades selling the world on the promise of a rules-based international order—open markets, integrated supply chains, multilateral cooperation—and now it's weaponizing that very integration against its own allies. The subordination Carney is talking about isn't conquest or occupation; it's the subordination of being excluded from a system you helped build, of watching America use economic interdependence as a cudgel rather than a collaboration.

"If we're not at the table, we're on the menu," he warned, and he wasn't just speaking for Canada. He was articulating the dawning realization of every middle power from the EU to ASEAN that America's pick me president has turned the entire post-war economic architecture into a performance stage for his own validation, and they're all just props in his psychodrama.

And what’s more terrifying than a pick me girl? A pick me president who just happens to be the president of the United States. Trump has a lot of power, especially in exploiting this loophole called the tariff (that’s still under Supreme Court review, but until the court rules he’ll keep playing around with it). It’s sort of funny how he brought the memo from Silicon Valley of “move fast and break things,” and break things he does. The problem is he’s breaking the entire American economic order in the process, like a pick me girl torching her female friendships to sit at the guys’ table, except the table is the global financial system.

It all started last year with the initial rounds of tariffs on Canada and Mexico around February of 2025. For Canadians, that devastated the auto sector, so much so that Stellantis and GM moved production out of the country entirely. The American share of car imports into Canada has been in free fall, now only accounting for about 39%, down from a high of 48% just a few years ago. Who’s filling that void? Mainly cars made in Korea and Japan. Congratulations, Trump—you played yourself.

Then you have the memorable April 2nd “Liberation Day.” I certainly felt very liberated that day seeing my net worth dropping by like 10% in the span of a week. The S&P 500 had its worst day since COVID, shedding over 10% as investors realized Trump wasn’t bluffing about economic suicide. The dollar index collapsed from 108 to 99 in a matter of weeks. I almost felt like capitalism is a scam and that I should embrace the purest form of communism a lot more. I was not only liberated, but also liquidated.

Xi didn’t like that, so he slapped Trump with a rare earth export blockage—you know, those materials essential for everything from iPhones to missiles. And Trump, being the coward that he is, backed down within 72 hours. There was a term that was coined after that: TACO. Trump Always Chickens Out. Carney learned that term too it seems, maybe from Xi after his most recent visit to China to patch things up. But more than getting a pat on the back from Xi, Carney will also bring back some Chinese EVs that just make American cars look like a complete joke. Those joint venture deals he’s negotiating? That’s Canada literally reshoring away from America.

I mean, if you don’t get Trump, at least understand the pick me girls. After all, they just want to be special, and they would do anything—like anything—to get attention. Like hanging out with six different male “friends” and loudly announcing how drama-free their life is compared to other girls. Trump is no exception because if there is anything he’s most scared of, it’s being forgotten. He felt that when Biden was president, overshadowed and irrelevant, and I am sure he didn’t like it very much.

So he, like the pick me girl, has to up his antics. Liberation Day last year was a big firework—markets crashing, the dollar tanking, 10-year Treasury yields spiking to levels not seen since the 2008 financial crisis as foreign investors dumped American debt. And now, for the second firework! Greenland! And I just checked the news—this man decided to back down from using military force to acquire Greenland. So yes, TACO. But hey, he’s a winner because once again everyone is talking about him. Gold prices hit another all time high, because he keeps scaring everyone with global stability on the line.

But honestly, he is a winner in his own way. This guy again distracted the American public from the Epstein case that was utterly disgraceful, and I don’t know how he keeps getting away with it. Also, he is once again dominating the media, everyone talking about him, me included, so he’s happy. He loves attention, remember? That’s the whole game.

But at what cost? At the cost of the American public who will probably never see the Epstein files, which is deeply troubling—how people in positions of power can be pedophiles and just... walk away. At the cost of American reputation, because no one believes what America says anymore, not even on paper. Treaties? Worthless. Promises? Jokes. And most importantly, at the cost of American economic interests, which is so funny because it was America who was into this whole reshoring thing, but now everyone is reshoring away from America.

It's sort of like when a pick me girl finally gaslights a man into being in a relationship with her. He showers her with attention and affection, and she's on top of the world. But at what cost? Well, she's happy in the moment, sure. But that happiness comes at the cost of every friendship she torched to get there, all those girls she threw under the bus to seem special.

And here's the cruel irony: the same desperate need for attention that made her burn those friendships doesn't just disappear once she gets the guy. She still craves that external validation, still needs to feel special. So when the relationship inevitably settles into routine and he's no longer showering her with the same intensity of affection—because that's what happens in actual relationships—she can't handle it.

She needs the fireworks, the drama, the spectacle. So she creates chaos, picks fights, makes scenes, anything to feel like the center of attention again. And when it all falls apart and they break up, she's left with nothing. Her reputation is in the bin, the girls she betrayed won't take her back, and no other men want to come close to her because everyone's seen this performance before. She burned her bridges for temporary validation, and in her addiction to that validation, she destroyed the very thing she fought so hard to get.

Trump's doing the same thing on the global stage—torching America's alliances (the girls) to feel powerful and dominant at home (the relationship), but his need for bigger and bigger spectacles means he can't even maintain the domestic support he craved, because eventually people get exhausted by the constant chaos.

In the real world, that pick me girl will always find another group of men and preach how she is different from other girls, how she’s not like them, how she’s better. But on the global stage, the world is only so big. And if you mess it up not once, but twice, no one will ever trust you again. Trust takes decades to earn, but once it’s lost, it almost never comes back.

Materially, this means trade flowing away from America towards China, evident with Carney’s recent deals with China that entail potential joint ventures to build Chinese EVs in Canada. Canadian travel to the U.S. in 2025 hit a record low. Or in the case of Europe, reducing the purchase of American assets and instead investing those assets into emerging markets or even China, which somehow is now considered a more reliable trading partner according to Carney. I actually laughed out loud at that one. The pick me girl is so insufferable that it has come to this—China is now the stable option. Let that sink in.

Longer-term, this could lead to a structural rise in the U.S. yield curve, which is already an issue as national debt continues to climb past $36 trillion. And Trump’s attacks on the Fed certainly don’t help—trying to bully Jerome Powell into rate cuts while simultaneously blowing up trade relationships is peak pick me behavior, demanding attention and accommodation while sabotaging the very foundation of your credibility.

So I do hate pick mes. Because they always believe they are different, special, somehow exempt from consequences. And somehow they surprise you in showing you that they truly are different—different in that they’re able to be out there selling out America just to prove that they’re special, to get validation, to stay in the headlines. Like, come on, what daddy issues are we working with here? They say and do things with an extent and shamelessness that just wows you. Seizing Greenland through military force? And they have no remorse, no regret, just onto the next stunt, the next headline, the next hit of attention.

But like any cockroach, the room is only so big, and you certainly cannot escape your stinky reputation. Macbeth found that out when the prophecy turned on him. The pick me girl finds out when she runs out of new friend groups to infiltrate. And Trump will find out when America runs out of global credibility to burn. The difference is, when a pick me girl crashes and burns, she only takes herself down. When a pick me president does it, he takes the empire with him.

Please note that the opinions expressed in this piece are my own and do not reflect those of my employers. Thanks for reading The Americano—don’t forget to share it with a friend!

The earnings calls came and went, as they always do. In the past week, JPMorgan Chase and Citigroup released their quarterly results, and the numbers told a familiar story. JPMorgan reported investment banking fees up 42%, driven once again by what Wall Street calls the “deal-making supercycle.” But buried beneath the headlines about capital markets and trading revenues was something more profound, something that struck me as I listened to executives discuss their technology investments. These institutions are not just banks anymore. They are becoming technology companies. And with that transformation comes a question that extends far beyond finance: What makes us human in the age of AI?

To understand what is happening, we need to start with what banks actually do. At its heart, banking is elegantly simple. Take deposits from people who put money in their checking and savings accounts, then lend that money out to individuals who need loans or credit cards, to families who want mortgages, and to businesses ranging from your neighborhood bagel shop to billion-dollar enterprises. Make sure you charge enough interest to cover your costs—paying interest on those deposits, paying your employees, keeping the lights on—and ensure that your clients can actually pay you back.

That last part is where things get complicated. The difficult reality of lending at scale is that clients do not always pay you back. We have seen it recently with the bankruptcies of companies like First Brands and Tricolor. But this is also where technology becomes remarkably efficient. Systems can perform Know Your Customer verification to assess a client’s risk and prevent fraud. Risk models can generate probabilistic outcomes on whether someone will repay a loan. And here is the beautiful mathematics of it all: when you run these operations on a large enough scale, using the same model, the law of large numbers works in your favor—unless, of course, your model is fundamentally flawed, in which case you are in serious trouble.

Think about it this way. Banks focused on lending are effectively becoming software companies. You write the code once, refine it, and then deploy it at scale with margins that improve as you grow. The lending itself can theoretically expand without limit, while the cost of running the model remains relatively fixed. It is a simplification, certainly, but it captures how the industry has evolved.

Companies have already proven this model. SoFi operates as an entirely digital bank with a banking charter. Chime markets itself as an all-in-one banking app with no monthly fees, no overdraft charges, and no minimum balance requirements. They are not exceptions anymore; they are the new normal.

Consider this thought experiment. It is the early 2000s, and you want to take out a personal loan for law school. You drive to your local bank branch, sit down with an advisor, fill out forms, and submit stacks of paperwork. Someone performs the underwriting manually, combing through your documents, assessing your creditworthiness. The process takes days. When everything is said and done, you are looking at a week or two before you hear back about your loan.

Now fast-forward to today. You open your laptop, fill out an online form in thirty minutes, submit it, and wait a few hours. The system processes your application, runs it through algorithms that assess thousands of data points, and returns a decision. Your credit card or loan approval arrives before dinner. The transformation is not just about convenience, though that matters. It is about efficiency, reduced risk for the bank, and dramatically lower costs. Hiring and training employees is tremendously expensive for banks. Automation changes the equation entirely.

The current chapter of this story involves generative AI, and it is playing out across every division of traditional banking. Trading floors have been in structural decline for years as algorithms replace human traders. Asset and wealth management are being transformed. Investment banking and corporate banking are next. The scale of this shift is staggering.

Recently, JPMorgan Asset Management, which manages more than $7 trillion, announced it is replacing proxy advisers with its AI system called Proxy IQ. Proxy advisers are specialized firms that help institutional investors decide how to vote their shares at corporate shareholder meetings, essentially guiding trillions of dollars in voting power. The new tool will aggregate and analyze proprietary data from over 3,000 annual meetings, eliminating the need for external advisors. The bank stated this transition allows it to vote solely in clients’ best interests. It is the first major investment firm to make such a move.

This is just the tip of the iceberg.

Beneath the surface of these corporate announcements, a new generation of startups is reimagining what it means to work in finance. Consider Rogo, a company with the audition ambition of becoming what its founders call an “AI-native investment bank.” The company was co-founded by Gabriel Stengel, who worked at Lazard, and John Willett, who worked at JP Morgan and Barclays, former investment bankers who experienced firsthand the brutal hours junior analysts spend on manual research, building financial models, and crafting presentation decks.

Rogo raised $50 million in a Series B round, bringing its total funding to $75 million. The platform integrates into bankers’ daily tools, primarily Excel, and automates tasks that traditionally consumed the lives of first-year analysts. It can draft memos, build models, create slides, and conduct research, all the “grunt work” that banking analysts endure during their hundred-hour work weeks. The promise is that human bankers can focus on what supposedly distinguishes senior professionals: client relationships and strategic thinking.

What Rogo and companies like it are doing is effectively eliminating the entry-level positions at investment banks and private equity firms. The work that remains is analyzing deals to determine whether they will generate returns, and the sales work where managing directors source opportunities. Those human elements, judgment and relationships, are not being replaced anytime soon. But the traditional path of paying your dues as a junior analyst, learning through exhaustion and repetition, is vanishing.

The employment data hints at the deeper transformation underway. At Harvard’s MBA program, only 4% of students received no job offer within three months of graduation in 2021; by 2024, that figure swelled to 15%. MIT saw a similar change, with its share of offer-less graduates climbing from 4.1% to 14.9% in just three years. And while experts suggest that an AI-fueled finance job takeover is largely “smoke and mirrors” for now, the direction is clear. A report from Citigroup found that 54% of financial jobs have a high potential for automation, more than any other sector.

So if we strip away all the work that AI can do better and faster than humans, what is left for us? I believe there are two essential capacities that remain distinctly human, at least for now.

The first is critical thinking, not the buzzword version taught in corporate training sessions, but the real thing. The ability to analyze an issue from multiple angles, gather facts and evidence from imperfect sources, synthesize them into a coherent understanding, reach a conclusion, and then act on it with conviction. This is where you see diverging performance between funds and banks. It is what we call strategy and execution.

The stock market offers a fascinating case study. Right now, investors are grappling with what some call the AI bubble, and it is remarkable to watch sophisticated people with access to the same information reach completely opposite conclusions.

Some investors see warning signs everywhere. They compare the current investment cycle to 1997, right before the dot com bubble. They point to stretched multiples on the S&P 500 index. They note OpenAI’s spending commitments exceed $1 trillion despite generating only a few billion in revenue. When DeepSeek released its model, these skeptics argued that large language models do not need as much compute power as previously imagined, potentially undermining the entire thesis behind data center investments. Prominent figures like Howard Marks have suggested that while the technology itself may be transformative, the front-loading of spending will likely create a bubble.

On the other hand, many investors, myself included, believe the investment in AI is far from a bubble. We see rapid advancement in AI training, scaling of models, and increased specialization in fields like drug research and autonomous driving. The technology is expanding into images, voice, and video. We point to structural bottlenecks that actually constrain enthusiasm — limited supplies of high-bandwidth memory, the challenge of building data centers and powering them. If anything, we argue, the addressable market is larger than currently imagined, and current spending remains underwhelming relative to the opportunity.

Both sides are smart people looking at the same data. What separates them is the ability to think through complex, ambiguous situations and make structurally sound decisions. This requires extensive learning and training in a domain. As Ken Griffin stated in a recent interview, AI can make short-term decisions in high-frequency trading, decisions made in seconds, but once you extend the timeframe, the models break down.

The second distinctly human capacity is building relationships. While many people hear “sales” and think of something negative—pushy, transactional, inauthentic—what it really means in high-stakes banking is creating genuine, trusting connections. When you need to get a deal done, you call someone you trust. When you are making a billion-dollar decision, you want to work with people you have built a relationship with over years.

This ability to build trusting, genuine working relationships is more difficult for AI to disrupt. I will not claim AI cannot eventually do it. In many ways, we do not even need AI for this disruption, ratings on Google Maps and Amazon already serve as trust validators. But in high-stakes banking, transactions require deep trust, and trust still comes primarily from human connection.

All of this extended reflection circles back to the fundamental question: What makes us human? Not in some abstract philosophical sense, but practically, in a world where machines can do more and more of what we once considered uniquely human work.

The answer, I think, is this: the ability to think freely and to be fully present with our senses, to imagine possibilities that do not yet exist, to connect with one another on a level deeper than transaction and efficiency. These capacities seem obvious when stated plainly, yet they are increasingly out of reach for many of us who grew up with smartphones and TikTok, who learned to seek answers in thirty-second clips rather than sustained thought, who built digital networks but struggle to build genuine, trusting relationships.

The great irony of our moment is that we have tools that could empower our lives in unprecedented ways, yet many people are becoming enslaved to them instead. They ask AI for answers rather than developing the capacity to think through problems themselves. They optimize for efficiency rather than meaning. They mistake productivity for purpose.

The transformation of banking offers a preview of what is coming for every industry. The routine work will be automated. The work that requires pure processing power or pattern recognition will be handled by machines. What remains will be the work that requires judgment, creativity, and human connection—the work that makes us most fully human.

There’s a moment in every economics student’s education when they encounter a curious omission. You spend an entire semester learning about production functions and trade policies. Then, in a single lecture, the professor mentions leisure. It’s important, they say, because it’s somehow the endpoint of all this economic activity. People working so they can eventually not work. Then the course moves on, and leisure is never discussed again.

I thought about this recently while scrolling through Instagram, double-tapping photos of places I’ll never visit, posted by people I’ve never met. My screen time report informed me I’d spent four hours on social media that day. Four hours. I couldn’t remember a single post.

This wasn’t always how we imagined abundance would feel.

In 1930, John Maynard Keynes made a prediction that would become one of economics’ most famous miscalculations. The father of modern macroeconomics forecast that technological progress would solve humanity’s economic problem within a century. By the 2030s, he wrote, we’d work perhaps 15 hours per week, leaving us to face what he called “his real, his permanent problem—how to use his freedom from pressing economic cares, how to occupy the leisure, which science and compound interest will have won for him, to live wisely and agreeably and well.”

Keynes got the productivity part right. He was spectacularly wrong about the leisure.

Today, as artificial intelligence promises to automate not just factory work but knowledge work itself, his question returns with uncomfortable urgency. If machines can think, what will humans do with all that time? But maybe we’re asking the wrong question. Maybe we should be asking: what happened to the time we already gained?

Between 1965 and 2003, according to research by economists Mark Aguiar and Erik Hurst, leisure time in the United States increased dramatically, 6 to 9 hours per week for men, 4 to 8 hours for women. We work less than our grandparents did. We have more free time than any generation in history. Yet we don’t feel leisurely. We feel exhausted, distracted, and somehow always behind.

Walk through any major airport and you’ll see the leisure economy’s split personality on full display. At one gate, travelers hunch over phones, scrolling through TikTok and Instagram, killing time before boarding. At another, a couple boards a first-class cabin that cost more than most people’s monthly rent, heading to an exclusive resort where every moment has been curated for maximum relaxation per dollar spent.

These aren’t just different choices. They’re symptoms of the same condition. Leisure has become a commodity, something to be purchased, optimized, and extracted for value. And like any commodity in a market economy, it can be made cheap or expensive, but it’s increasingly difficult to find it simply... free.

Meta’s 2024 earnings report provides one measure of this transformation. The company generated $164.5 billion in revenue, up 22% from two years earlier, extracted from 3 billion daily active users across Facebook, Instagram, and WhatsApp. The business model is elegant in its simplicity: the longer users stay on the platform, the more ads the company can serve. Every minute of leisure time captured translates directly to shareholder value. Mark Zuckerberg isn’t in the social media business, really. He’s in the leisure arbitrage business, buying your time at the price of free entertainment and selling it to advertisers at premium rates.

On the other end of the spectrum, luxury has never been more luxurious. Air France recently invested hundreds of millions upgrading its first-class cabins. Ultra-luxury hotel chains like Aman and Rosewood are expanding rapidly, with nightly rates that can exceed $2,000. Companies offering bespoke Japanese travel experiences are preparing for IPOs. If you have money but no time, you can purchase efficiency—experiences designed to deliver maximum restoration in minimum hours.

Both approaches treat leisure as a production problem. The only question is whether you’re optimizing for volume or quality, cost or exclusivity. What’s disappeared is the notion of leisure as something that exists outside the logic of markets entirely.

There’s an economic framework that helps explain this, though it’s rarely presented as the dystopian diagnosis it implies. The Beckerian model of household production treats leisure not as rest but as manufacturing. Households become little factories, combining “leisure capital”—televisions, smartphones, streaming subscriptions, gaming consoles—with time to produce leisure services. As leisure capital has gotten cheaper (a Netflix subscription costs less than a single movie ticket in 1950s dollars), households can substitute capital for time, extracting more leisure per hour.

This explains a paradox that puzzles many people: why do high-income workers often work more hours than low-income workers, even though they can afford more leisure? Because their time is expensive. A lawyer billing $500 an hour has a strong incentive to purchase leisure capital—the premium gym membership, the meal delivery service, the expensive vacation—that maximizes restoration per hour, freeing up more time to bill more hours. It’s not that they don’t value leisure. They value it so much they’ve turned it into an investment.

The problem reveals itself in the data on how we actually spend this leisure we’ve gained. Between 2014 and 2024, the percentage of Americans who socialize with friends on a typical day fell from 38% to 30%, according to the American Time Use Survey. We’re not using our free time to deepen relationships or build communities. We’re using it to consume content alone.

In 2023, the U.S. Surgeon General declared loneliness a public health epidemic, noting it carries health risks equivalent to smoking 15 cigarettes a day. We have more ways to connect than ever—more platforms, more apps, more options—yet we’re more isolated. Leisure time has increased, but what social scientists call “social capital”—the sympathy, empathy, and shared commonalities that facilitate genuine cooperation—has declined.

What replaced it might be called cheap social capital: the likes, comments, shares, and group chats. Low-effort digital substitutes that provide the appearance of connection without its substance. It’s the social equivalent of fast food—engineered for maximum palatability and minimum nutritional value.

There’s a historical pattern here that bears an uncomfortable resemblance to other economic transformations. In 18th-century England, common lands, spaces shared by communities for grazing, gathering, and recreation, were enclosed and privatized. This created the conditions for market economies but destroyed communal ways of life. What was once shared became property. What was once free became priced.

Today’s enclosure operates in the cultural sphere. Non-commodified spaces, human relationships, spontaneous joy, unstructured time, are being fenced off and transformed into products for consumption. Social networks enclose friendship. Dating apps enclose romance. Mindfulness apps enclose meditation. Everything that was once freely shared now requires a subscription.

The fertility data suggests this logic has reached even our most intimate choices. Across developed nations, birth rates have collapsed: 1.64 births per woman in the United States, 1.33 in Italy, 1.26 in Spain, 0.72 in South Korea. While multiple factors drive this decline, the economic reframing of family deserves attention. When leisure becomes a commodity to be optimally consumed, children transform from a dimension of human experience into a utility calculation. The time cost, the financial cost, the opportunity cost, and suddenly family formation becomes a question of individual optimization rather than something people do because it’s meaningful.

I think about this when I see my friends, most of whom are in their twenties, talking about whether they can “afford” to have kids. They don’t mean it purely financially, though that’s part of it. They mean: can I afford the time? What will I have to give up? What experiences will I miss? These are reasonable questions in an expensive world. But they’re also questions that would have seemed bizarre to most humans throughout history, for whom children weren’t a choice on a menu of lifestyle options but simply what came next.

Technology and capitalism promised to free us from toil. Instead, they’ve invaded every corner of life, including the parts that were supposed to be refuges from market logic. We’ve imported the vocabulary of productivity into relationships (networking), self-care (optimization), and even rest (recovery). Nothing is allowed to simply be. Everything must justify itself in terms of output.

As AI accelerates this transformation, promising to automate even more of our working hours, Keynes’s question returns: what will we do with all that time? But the answer seems increasingly clear. We won’t spend it in contemplation or community. We’ll spend it alone, staring at screens, while algorithms optimize our attention for maximum advertising revenue. Or, if we’re wealthy enough, we’ll purchase curated experiences designed for maximum efficiency, treating leisure itself as another form of work.

The great irony of our age is that we’ve achieved Keynes’s dream of abundance, but we let it be colonized before we could inhabit it. Technology gave us more time. Economics taught us to optimize it. And now, immersed in infinite content and exclusive experiences, we rarely stop to ask the most basic question: leisure for what?

The answer won’t be found in economics textbooks or productivity apps. It requires remembering something we knew before leisure became a market—that time spent with others for no particular reason, activities pursued for their own sake, and connections formed without transaction aren’t inefficiencies to be optimized away. They’re the point.

Until we reclaim leisure from commodification, we risk becoming the wealthiest, most entertained, and loneliest civilization in human history. We will have gained all the time in the world and lost the meaning of it.

Happy holidays! We hope you’re spending this time surrounded by joy and festivity, with your loved ones.

2025 has been a turbulent year. President Trump took office, and many of his policies have thrown the world into chaos, including America’s continued pressure on Venezuela. We also witnessed the escalation of war in the Middle East involving Israel and Iran before it subsided, as well as the ongoing conflict between Russia and Ukraine and tensions that persist in Asia, including in Nepal.

For many of us, what hit closest to home was the tough labor market for those entering the workforce, the constant fear of AI replacing our jobs, concerns about an AI bubble erasing equity market gains, inflation, and the rising cost of food—with record prices for eggs and orange juice. All of this is to say: the world has been turbulent.

Despite all of this, you still took time out of your day to read the Americano, and for that we are tremendously grateful. We thoroughly enjoyed reading feedback from so many of you (you know who you are). Your responses have been nothing but thoughtful, humorous, and at times, have provided the much-needed criticism that challenges our viewpoint and pushed us to explore new dimensions of the same issue or to discover topics we didn’t know existed altogether. Please continue to share your feedback and thoughts, and share any pieces you find relevant with your friends and colleagues.

To share a bit more about the Americano’s year-end wrap: We’ve had the chance to have conversations with many standout individuals in their respective fields, including Dr. Lindell, a computer vision scientist, and Mike Bird, the Finance and Economics editor at the Economist, as well as countless creators like Lilian Zhang and business owners like Sarah Abell and Alex Cronin.

The Americano’s most popular piece was “Do Women Hate Men Now,” which deeply reflected the state of division between young men and women and how impossible it feels to date anyone. The second most popular piece was “Breaking Down the $50 Jellycat: What You’re Actually Paying For,” as we can all deeply relate to the desire to collect more plushies.

As for 2026, the Americano will continue as usual but with a few changes. First, instead of sticking with the 5-times-a-week publication cycle, we’ll be trimming it to 3 pieces a week: Monday, Wednesday, and Friday. This change in our publication cycle will allow our team to focus on quality research and provide the depth and engagement you’re looking for.

Second, we’re looking forward to hearing from readers who’d like to share their experiences with the economy and address their concerns. This could include questions like “Should I switch jobs every two years?” or “Should I buy Bitcoin or gold?” We want to hear more from you.

Third, we’re also looking for guest submissions if you want to voice your concerns about particular economic issues or share insights from working in a particular industry. Many of our readers are prolific writers eager to have their voices heard, and we would love to read about it, so please send it in.

Looking ahead to 2026, we have many amazing stories planned. The first major piece we want to cover relates to housing and all of your questions regarding affordability, and the role of homeownership in shaping political and social viewpoints on issues like marriage. If you’re interested in or know someone who would like to participate or contribute to this story, please reply to this email.

In addition, we’re also planning to feature more guest interviews (if you have someone you think we should talk to, drop their name) and host events, including a book talk with Mike Bird about his recent release, The Land Trap, which extensively discusses the financialization of land.

Lastly, a special thank you for choosing to be a paid subscriber. Beyond the monetary contribution in supporting our work, your subscription is really a vote of confidence in our mission and the stories we want to tell.

Happy Holidays,

The Americano Team

You’ve probably thought about it. Quitting your corporate job or dropping out of school to start that coffee shop, launching your own clothing line, or finally building that app idea you’ve been sitting on. The dream feels so close. It is just one leap away from freedom, purpose, and doing something that’s actually yours.

You’re not alone. According to a 2020 survey from WP Engine, 62% of Gen-Z plan to start or have already started their own business. Your Instagram feed is full of 25-year-olds running seven-figure businesses from Bali, and TikTok entrepreneurs promise you can dropship your way to financial freedom.

But here’s what nobody’s telling you: most of those stories on social media are leaving out the parts that matter most. The behind-the-scenes reality looks nothing like the highlight reel. So what does running a small business actually look like?

I recently had the chance to talk to Alex Cronin. Approaching 40, he’s built something real. His company, Freetone Printing, has printed apparel for Heineken, Audi, and Mejuri, and they’re on track to gross over $1 million in revenue this year with just three people. On the side, his clothing brand Permanent Vacation generated an estimated $250,000 in sales this year.

To any bystander scrolling through social media, this sounds like the dream. What people don’t show you is that Alex nearly lost everything getting here.

Alex started with Permanent Vacation, a streetwear brand built on uncompromising quality. He sourced premium fabrics and obsessed over every detail. The brand became a slow burn—building a loyal following, establishing credibility, creating something he was genuinely proud of.

Then March 2020 hit, and everything collapsed.

When the pandemic shut down global supply chains, the specific fabrics Alex used couldn’t be shipped to Canada anymore. He faced a choice: switch to lower-quality fabrics and keep revenue flowing, or stop selling altogether.

Alex chose to stop. His income went to zero in a matter of weeks.

“I refused to compromise on quality,” he told me. But principle doesn’t pay the bills. Alex took out loans from banks. He borrowed money from family and friends. He did whatever he could to stay afloat, drowning in debt with no clear path forward.

Imagine going to your friend in your 30s and telling them you need to borrow money. Asking the question is already hard enough, but asking it several times is hell. That’s the part that

gets skipped over. That’s the reality that 62% of Gen-Z aspiring entrepreneurs have no idea they might be walking into.

The turning point came from Alex’s frustration with the poor quality of printing services for Permanent Vacation. After other printing companies kept messing up his apparel, Alex made a desperate decision: buy the machinery and do it himself.

Think about how insane that decision must have seemed. Alex was already drowning in debt from a failing clothing company. The logical thing would have been to cut losses and get a stable job. Instead, he borrowed more money to buy industrial printing equipment and learn an entirely new skill set from scratch.

When I asked what it took to learn those machines, he gave me a simple answer: “If you have a curiosity, then you have the patience to learn it.”

While his friends were enjoying their free time after their 9-to-5s, Alex was teaching himself screen printing, troubleshooting machines that cost more than many people’s annual salary, betting everything on a pivot that might not work.

A few years in, through slow and steady work, the printing business became viable. The same obsessive attention to quality that nearly bankrupted Permanent Vacation became Freetone Printing’s competitive advantage. In a market full of cheap, fast printing services, they became known for doing it right.

Alex’s story matters because it reveals something crucial about the entrepreneurship craze among Gen-Z. We’re not just chasing a dream—we’re fleeing a nightmare.

Gen-Z is facing a fundamentally different economic reality than previous generations. According to recent surveys, 56% of Gen-Z doing independent work would actually prefer a permanent, stable job. More than half of us who are “choosing” entrepreneurship would rather have traditional employment if it were actually viable. We’re working multiple jobs not out of ambition but necessity. The gig economy isn’t freedom—it’s survival dressed up in startup language.

McKinsey, a consultancy firm, found that 59% of Gen-Z don’t expect to ever own a home, as the average age of homeownership climbs to 56. Most of us will be approaching retirement age before we can afford what our grandparents bought in their twenties. When the traditional markers of success feel impossible, starting your own business starts to look less like ambition and more like the only path left.

Into this void stepped an army of influencers selling escape. Gen-Z grew up with social media portraying life as rosier than reality, from Tai Lopez’s garage Lamborghini to today’s laptop-on-a-beach entrepreneurs. The formula is always the same: show the lifestyle, hint at vague struggles, then sell you their system. When you’re 25, working a job you hate, watching your paycheck disappear into rent and student loans, those stories become magnetic. But that version is dangerously incomplete, missing the bankruptcy and sleepless nights.

Our generation also desperately craves purpose in a world that seems designed to strip it away. We grew up in a digital world where life became frictionless to the point of feeling meaningless. We’re more pessimistic and depressed than previous generations. Religion has declined sharply, leaving many without the frameworks that gave previous generations meaning.

Entrepreneurship promises to fill that void. It offers purpose, autonomy, and the feeling that you’re building something that matters. The problem is that when you’re running from economic desperation while seeking existential purpose, you’re vulnerable to anyone selling an easy solution.

I asked Alex what he’d tell younger people considering entrepreneurship. His answer wasn’t what most people want to hear.

“People try to skip crucial steps because they hear over-glorified stories,” he said. “You should take a few years to work and learn. Inspiration exists, but it has to find you working.”

Alex didn’t start his successful business in his 20s—he’s almost 40 now. And that’s actually the norm, not the exception. Research consistently shows that most successful entrepreneurs start their businesses in their 30s and 40s, not their 20s. They succeed because they’ve spent decades building skills, understanding industries, and accumulating knowledge that can only come from experience.

But it’s more glamorous to cover Mark Zuckerberg dropping out of Harvard at 19 or Bill Gates starting Microsoft at 20. These stories are real, but they’re extreme outliers. For every 20-something founder who makes it, there are thousands who crash and burn, and we never hear about them because failure doesn’t go viral.

“Try different things,” Alex told me. “You don’t have to quit your job, but do consider working in an industry that interests you.”

This is the advice that doesn’t sell courses or get likes on LinkedIn, but it’s what actually works. Build skills in your 20s. Make mistakes on someone else’s dime. Learn how businesses actually operate. Build relationships. Save money. Develop expertise. Then, when you’re ready—when you have the knowledge, the network, the financial cushion, and the resilience—start something of your own. This is how Stephen Schwarzman founded Blackstone, now the largest private equity fund, and how Vera Wang started her own prominent wedding dress brand after working at Vogue for 17 years.

Social media has made success look so quick and attainable that we’ve forgotten how long real mastery takes. Alex’s journey shows that building something real takes time, sacrifice, and often comes at a cost nobody warns you about.

What struck me most is that despite everything—the near-bankruptcy, the stress, the loans, the years of uncertainty—Alex wouldn’t trade his journey for anything. What kept him going was his love for the business itself. Not the idea of being an entrepreneur. Not the lifestyle or freedom. The actual work.

That’s the difference between entrepreneurship as escape and entrepreneurship as calling. And it’s a difference that determines whether you survive the inevitable moments when everything falls apart.

For Gen-Z, you shouldn’t start a business because you think it will magically give you purpose, or because you believe entrepreneurship is a quick way to strike gold. You should start a business because you’re so obsessed with solving a specific problem or creating a specific thing that you’re willing to nearly lose everything for it.

The entrepreneurship dream is real. But so is the nightmare of nearly losing everything. Nobody on Instagram will show you the loans from family, the panic attacks at 3 AM, or the years of grinding it takes—way more years than you think—with no guarantee any of it will work out.

Alex’s success didn’t come from following a formula or taking a leap of faith. It came from curiosity strong enough to generate patience, from refusing to compromise on quality even when it nearly destroyed him, and from loving the actual work enough to keep going when any sane person would have quit.

That’s the reality. Not the highlight reel.

This is the contradiction that nearly destroyed Patagonia’s brand: How do you maintain authenticity as an environmental activist company when your biggest customers are the very corporations driving climate change?

But here’s what makes Patagonia fascinating: the company actually meant it. And in 2022, they proved it in the most radical way possible.

Yvon Chouinard, the 83-year-old founder worth billions, gave the entire company away. Not sold it. Not took it public. Gave it away. To fight climate change. In doing so, he created a blueprint for what authentic corporate activism actually looks like in an era where everyone claims to care but few actually do anything about it.

The story starts in 1957, when a 19-year-old Yvon Chouinard bought a used coal-fired forge from a junkyard because he was frustrated with climbing gear. He wasn’t trying to build a company. He wasn’t following market trends. He was a climber who needed better equipment, so he taught himself blacksmithing and started making chrome-molybdenum steel pitons—those metal spikes you hammer into rock faces. They were lighter and stronger than anything else available, and he sold them for $1.50 each from the back of his car.

By 1965, Chouinard Equipment had become the largest supplier of climbing hardware in America. Success. Profit. Growth. Everything you’re supposed to want as an entrepreneur. Then Chouinard went climbing in Yosemite and realized his own pitons were destroying the rock faces. The scars were permanent. His product was killing the planet he loved.

So in 1972, he did something almost unthinkable in business: he stopped making the product that built his company. Instead, he introduced aluminum chocks—removable gear that didn’t damage the rock—and published an essay called “The Whole Natural Art of Climbing” that essentially told climbers to stop using his most profitable product. This launched the “Clean Climbing” movement and established something crucial about the company that would become Patagonia: it would sacrifice profit for environmental integrity. Not as marketing. As a principle.

The clothing business started by accident. Chouinard brought back a rugby shirt from Scotland in 1970 because the heavy collar prevented climbing slings from cutting into his neck. His friends wanted them. So he started importing rugby shirts, rain gear, and wool gloves—not because he wanted to be in fashion, but because climbers needed functional clothing. In 1973, he named the clothing line Patagonia, choosing a word that evoked remote wilderness and worked in multiple languages.

What followed was decades of material innovation driven not by trends but by solving real problems. In the late 1970s, Patagonia discovered synthetic pile fabric that had been discarded after the fake-fur market collapsed. They turned it into Synchilla fleece—the material that still defines outdoor gear today. In 1980, they introduced polypropylene base layers, originally used in marine ropes and diaper linings, because it didn’t absorb water. They developed the three-layer system—moisture transport, insulation, shell—that fundamentally changed how people dressed for extreme conditions.

None of this was about fashion. It was about function. And that functional authenticity is what built the brand’s reputation.

But by the 2010s, something unexpected happened. The trend of wearing high-performance outdoor gear in urban settings—dubbed “Gorpcore”—turned Patagonia into a status symbol. Those Synchilla fleeces and puffer jackets designed for Patagonian glaciers became the uniform of Wall Street and Silicon Valley. The brand that stood for environmental activism became associated with the exact people whose industries were driving climate change.

This wasn’t just awkward—it was existential. Patagonia’s entire brand identity was built on authenticity, on doing the right thing even when it hurt profits, on fighting climate change. And now their customer base included hedge fund managers funding fossil fuel expansion and tech executives building energy-intensive data centers.

Patagonia tried to address this tension through action rather than rhetoric. They launched Worn Wear in 2017, encouraging customers to repair gear instead of buying new products. They operate the largest garment repair center in North America in Reno, Nevada, fixing over 40,000 items in 2024 alone. They created a trade-in program offering store credit for used gear. They ran ads on Black Friday saying “Don’t Buy This Jacket,” urging customers to consume less.

But the association persisted. So Patagonia took an unusual step: they ended corporate sales to certain financial and consulting firms. They turned away revenue from some of the wealthiest corporate clients in America because those clients didn’t align with their mission.

Consider what that means. A company choosing principle over profit in a way that actually costs money. That’s rare enough to be noteworthy.

But the real turning point came in September 2022, when Yvon Chouinard restructured ownership of the entire company in a way that shocked the business world. He transferred 100% of Patagonia to two entities: the Patagonia Purpose Trust, which holds all voting stock to protect the company’s mission, and the Holdfast Collective, a 501(c)(4) that holds all non-voting stock and receives 100% of profits not reinvested in the business.

What that means in practice is that every year, after Patagonia covers operating costs and reinvestment, all remaining profits—roughly $100 million annually—go to the Holdfast Collective to fight climate change. Not as charitable donations. As the company’s actual purpose. Patagonia exists to generate money for environmental activism. The clothing is the means, not the end.

Since August 2022, Holdfast has received approximately $180 million from Patagonia. That money is funding lawsuits against fossil fuel companies, supporting indigenous land protections, and lobbying for climate legislation. Patagonia sued the Trump administration over protected lands. They close stores on Election Day to encourage voting. They’ve consistently backed their stated values with legal and financial action.

This approach appears to work. By fiscal 2025, Patagonia sourced 84.4% of materials from preferred sources with reduced climate impacts. They eliminated PFAS “forever chemicals” from waterproof shells. They created the NetPlus line from recycled fishing nets. They’re not perfect—no company operating at scale can be—but the commitment is measurable and documented.

I can attest to their repair commitment personally. I own a Patagonia down jacket that survived four Toronto winters. After four years of daily wear, the entire back was worn through—fabric torn, insulation exposed. I brought it to the store expecting to be told I needed a replacement. Instead, they sent it to their repair center. A month later, it came back fully repaired. A lesser-known feature: you can request different colored patches to customize repairs. My jacket now has mismatched patches that make it distinctly mine. I haven’t encountered another brand offering anything close to that level of lifetime warranty commitment.

But this success created its own challenge. As Patagonia became synonymous with environmental consciousness, people started wearing it not because they shared those values, but because they wanted to signal that they did. The finance professionals buying Patagonia weren’t necessarily looking for gear that would last a lifetime. They were buying into an image while working for companies whose business models contradicted that image.

That’s why Patagonia’s decision to sever ties with those corporations mattered. They chose integrity over revenue. And for consumers increasingly skeptical of corporate activism, that choice carries weight.

Because the broader market context makes Patagonia’s approach unusual. Every brand now claims to care about sustainability. Every corporation has sustainability officers and ESG reports. Every CEO discusses stakeholder capitalism. But when environmental commitments conflict with profit, most companies quietly backtrack.

Patagonia hasn’t. They’ve demonstrated this repeatedly over 50 years. When their pitons damaged rocks, they stopped selling them. When fleece production had environmental costs, they switched to recycled materials. When their customer base conflicted with their values, they turned away revenue. When Yvon Chouinard could have sold for billions, he gave it away instead.

That track record is why the brand maintains credibility despite the finance vest association. The company has consistently chosen principle over profit. Not as marketing strategy. As operational practice.

The broader lesson here isn’t just about Patagonia. It’s about authenticity in corporate messaging. Consumers—particularly younger demographics—can distinguish between companies that claim values and companies that operationalize them. They notice when sustainability reports contain vague commitments without concrete timelines. They notice when companies celebrate social causes publicly while funding opposing interests. They notice when ethical claims are branding exercises rather than business practices.

Patagonia works because the commitments are verifiable. The repairs are real. The material sourcing is documented. The financial restructuring is public record. The policy advocacy is on the record. They’re not perfect—perfection isn’t possible at scale—but they’re demonstrably aligning their business model with their stated mission, even when it reduces revenue.

In a market where every brand attempts to appeal to conscious consumers through performative messaging, verifiable authenticity becomes valuable. It’s also difficult to replicate, which is why few companies achieve it.

The finance professionals almost derailed Patagonia’s brand identity. But the company’s response—severing those relationships, reinforcing their mission, restructuring ownership to ensure profits fund climate action—demonstrated that their commitment was operational, not rhetorical. That’s why, despite the mockery and the fleece vest jokes, Patagonia maintains its reputation among consumers who trust few corporate brands.

They earned it through consistent action rather than marketing campaigns. Through 50 years of choosing values over profits when those choices conflicted, even when the decision wasn’t publicly visible.

That kind of authenticity can’t be purchased through advertising. And in a market saturated with performative corporate messaging, it’s the only kind that maintains credibility.