AI

Kevin Warsh Channels Alan Greenspan in AI Productivity Bet

Published

5 months ago

February 5, 2026

When Kevin Warsh steps into the ornate confines of the Federal Reserve’s Eccles Building—assuming Senate confirmation—he’ll carry with him a wager that could define the American economy for a generation. Donald Trump’s nominee for Fed chair is betting that artificial intelligence will unleash a productivity boom powerful enough to justify aggressive interest rate cuts without reigniting inflation, echoing the audacious gamble Alan Greenspan made during the internet revolution of the 1990s.

It’s a high-stakes proposition. Get it right, and Warsh could preside over an era of robust growth and falling prices reminiscent of the late Clinton years. Get it wrong, and he risks stoking the very inflation demons the Fed has spent years battling. As economists debate whether AI represents the most productivity-enhancing wave since electrification or merely another overhyped technology cycle, Warsh’s nomination has become a referendum on America’s economic future.

Echoes of the 1990s: Greenspan’s Legacy Revisited

The parallels to Greenspan’s tenure are striking—and deliberate. In the mid-1990s, as the internet began reshaping commerce and communication, mainstream economists warned that the US economy was overheating. Unemployment had fallen below 5%, traditionally considered the threshold for accelerating wage growth and inflation. The conventional playbook called for rate hikes to cool demand.

Greenspan defied orthodoxy. Convinced that internet-driven productivity gains were fundamentally altering the economy’s speed limit, he held rates steady and even cut them in 1998. The gamble paid off spectacularly: productivity growth surged from an anemic 1.4% annually in the early 1990s to 2.5% by decade’s end, while core inflation remained tame. The economy expanded at a 4% clip, unemployment fell to 4%, and the federal budget swung into surplus.

Now Warsh appears poised to replay that script with AI as the protagonist. In a Wall Street Journal op-ed last year, he described artificial intelligence as “the most productivity-enhancing wave of technological innovation since the advent of computing itself.” His thesis: AI will drive down costs across the economy while supercharging output, creating a disinflationary force that allows the Fed to maintain easier monetary policy without courting price instability.

The timing is provocative. After hiking rates from near-zero to over 5% to combat post-pandemic inflation, the Fed under Jerome Powell has adopted a cautious stance. But recent data suggests Warsh may have identified an inflection point: productivity growth has accelerated to 2.1% annually, according to calculations by The People’s Economist, while inflation has cooled to near the Fed’s 2% target. Meanwhile, corporate America is pouring unprecedented capital into AI infrastructure—Google parent Alphabet alone has committed $185 billion over several years to AI data centers and computing capacity.

The AI Productivity Wager: Data and Doubts

Yet the AI productivity bet rests on assumptions that many economists find uncomfortably optimistic. While Greenspan could point to visible productivity gains from internet adoption—e-commerce, email, digital supply chains—AI’s economic impact remains largely theoretical.

Consider the evidence on both sides of this consequential debate:

The Optimistic Case:

Investment tsunami: Big Tech companies have announced over $500 billion in AI-related capital expenditure through 2027, potentially eclipsing the infrastructure buildout of the internet era
Early productivity signals: Goldman Sachs research suggests AI could boost US labor productivity growth by 1.5 percentage points annually over the next decade
Deflationary mechanisms: AI-powered automation is already reducing costs in customer service, software development, legal research, and medical diagnostics
Broad applicability: Unlike previous technologies limited to specific sectors, AI promises productivity gains across virtually every industry from agriculture to healthcare

The Skeptical Counterargument:

Implementation lag: As The Economist notes, productivity gains from transformative technologies typically take 10-15 years to materialize fully—Greenspan’s bet benefited from fortuitous timing as gains accelerated just as he cut rates
Measurement challenges: Productivity statistics notoriously struggle to capture improvements in service quality, potentially understating gains but also making real-time policy decisions hazardous
Displacement costs: AI-driven job disruption could create transitional unemployment and reduce consumer spending, offsetting productivity benefits
Energy demands: AI data centers consume massive electricity, potentially creating inflationary pressure in energy markets that could offset disinflationary effects elsewhere

The comparison between the 1990s internet boom and today’s AI surge reveals both similarities and critical differences:

Metric	1990s Internet Era	2026 AI Era
Productivity Growth	1.4% → 2.5% over decade	1.5% → 2.1% (18 months)
Capital Investment	~$2 trillion (inflation-adjusted)	Projected $500B+ through 2027
Inflation Environment	Stable 2-3% range	Recently peaked at 9%, now ~2%
Fed Funds Rate	Gradually lowered from 6% to 5%	Currently 5.25-5.5%, pressure to cut
Adoption Timeline	15+ years to mass adoption	Rapid deployment but uncertain ROI
Labor Market	Unemployment fell to 4%	Currently 3.7%, near historic lows

Desmond Lachman of the American Enterprise Institute offers a sobering caution in Project Syndicate. While acknowledging Warsh’s qualifications to navigate the AI revolution, Lachman warns that premature rate cuts could spook bond markets, particularly given elevated government debt levels that dwarf those of the 1990s. Federal debt stood at 60% of GDP when Greenspan made his bet; today it exceeds 120%.

Implications for the US Economy and Growth Trajectory

The stakes extend far beyond monetary policy arcana. Warsh’s AI productivity bet carries profound implications for workers, businesses, and America’s competitive position.

If AI delivers on its promise as a disinflationary force, the US economy could enter a golden period of what economists call “immaculate disinflation”—falling inflation without the recession typically required to achieve it. Real wages would rise as nominal pay increases outpace price growth. The Fed could maintain accommodative policy, supporting business investment and job creation. Housing affordability might improve as mortgage rates decline. Stock markets, particularly growth-oriented technology shares, would likely soar on expectations of sustainably higher earnings.

But this optimistic scenario requires several conditions to align. First, productivity gains must materialize quickly—not in the usual decade-plus timeframe—to validate easier policy. Second, AI’s benefits must diffuse broadly across the economy rather than concentrating in a handful of tech giants. Third, labor market adjustments must occur smoothly without triggering political backlash that could derail the technological transition.

The risks of miscalculation loom large. As The New York Times editorial board cautioned, the Fed’s credibility—painstakingly rebuilt after taming inflation—could be squandered if premature rate cuts reignite price pressures. Workers on fixed incomes and retirees would suffer disproportionately. The Fed might then face the painful choice between tolerating higher inflation or hiking rates sharply enough to trigger recession.

There’s also the political dimension. Warsh’s nomination by Trump, who has repeatedly criticized Powell for maintaining restrictive policy, raises questions about Fed independence. While Warsh has a track record of intellectual autonomy—he dissented against some of the Fed’s crisis-era policies as a Governor from 2006-2011—the optics of a Trump-appointed chair cutting rates aggressively ahead of the 2028 election could undermine public confidence in the institution’s apolitical mandate.

Learning from History Without Repeating It

The Greenspan precedent offers both inspiration and warning. Yes, the Maestro’s productivity bet succeeded brilliantly—for a time. But his extended period of easy money also inflated the dot-com bubble that burst spectacularly in 2000, wiping out $5 trillion in market value. Critics argue his approach sowed the seeds of subsequent financial instability, including the housing bubble that culminated in the 2008 crisis.

Warsh, to his credit, has shown awareness of these pitfalls. As a Fed Governor during the financial crisis, he advocated for earlier recognition of asset bubbles and tighter oversight of financial institutions. His 2025 writings emphasize the need for “vigilant monitoring of financial stability risks” even as the Fed pursues growth-oriented policies.

The question is whether he can thread this needle—cutting rates to accommodate productivity gains while preventing the kind of speculative excess that characterized the late 1990s. The answer may depend less on economic theory than on judgment, timing, and some measure of luck.

The Verdict: A Calculated Gamble Worth Taking?

So is Warsh’s AI productivity bet sound policy or dangerous hubris? The honest answer is that we won’t know for several years, and by then the consequences—positive or negative—will already be unfolding.

What we can say is this: the bet is intellectually coherent, grounded in plausible economic mechanisms, and supported by preliminary data. AI does appear to be driving genuine productivity improvements, even if their ultimate magnitude remains uncertain. The disinflationary forces Warsh identifies—automation, improved resource allocation, reduced transaction costs—are real and observable.

But coherence doesn’t guarantee correctness. The 1990s productivity boom emerged from technologies that were already mature and widely deployed by mid-decade. Today’s AI tools, while impressive, remain in their infancy with uncertain commercial applications beyond a handful of use cases. The gap between technological potential and economic reality has tripped up many forecasters.

Perhaps the most balanced perspective comes from examining not just the economics but the political economy. A Fed chair’s primary job isn’t to achieve optimal policy in some abstract sense—it’s to maintain the institutional legitimacy necessary to conduct monetary policy effectively over time. That requires building consensus, communicating clearly, and preserving independence from political pressure.

On these criteria, Warsh brings both strengths and vulnerabilities. His intellectual firepower and private sector experience (he worked at Morgan Stanley before joining the Fed) command respect in financial markets. His youth—he’d be one of the youngest Fed chairs in history—signals fresh thinking. But his close ties to Trump and Wall Street could make him a lightning rod for criticism if his policies falter.

Conclusion: The Most Consequential Fed Chair Since Greenspan?

As Kevin Warsh prepares for confirmation hearings, he stands at a crossroads that could define not just his tenure but the trajectory of the US economy for decades. His AI productivity bet represents the kind of paradigm-shifting policy vision that comes along once in a generation—for better or worse.

If he’s right, future historians may rank him alongside Greenspan and Paul Volcker as transformational Fed chairs who correctly identified tectonic economic shifts and adjusted policy accordingly. We could be entering an era where technology-driven productivity gains allow faster growth with lower inflation, improving living standards across income levels while maintaining US economic dominance.

If he’s wrong, the consequences could range from merely embarrassing—a Fed chair who cut rates prematurely and had to reverse course—to genuinely damaging, with renewed inflation, financial instability, or the policy credibility erosion that made the 1970s such a painful decade.

The truth, as usual, likely lies somewhere in between these extremes. AI will probably deliver meaningful but not transformational productivity gains over the next 5-10 years. Policy will muddle through with some successes and some setbacks. The economy will neither enter utopia nor collapse.

But “muddling through” is an unsatisfying conclusion for an award-winning columnist to offer readers. So here’s a bolder prediction: Warsh will cut rates more aggressively than current market pricing suggests—perhaps 100-150 basis points over his first 18 months—justified by his AI productivity thesis. Growth will initially accelerate, validating his approach. But by 2028, signs of overheating will emerge—not in consumer prices but in asset markets, particularly AI-adjacent stocks and commercial real estate serving data centers. The Fed will face pressure to tighten, creating volatility.

The ultimate judgment on Warsh’s tenure will then depend on whether he shows the flexibility to adjust course when reality deviates from theory—something Greenspan struggled with in his later years. That capacity for intellectual humility and policy adaptation, more than the theoretical soundness of any particular bet, separates adequate Fed chairs from great ones.

For now, we can only watch, wait, and hope that Warsh’s AI productivity wager proves as prescient as Greenspan’s internet bet—without the bubble that followed.

Discover more from The Economy

Subscribe to get the latest posts sent to your email.

Up Next

Pakistan Must Create 30 Million Jobs Over the Next Decade or Face Instability, World Bank President Warns

Don't Miss

US Tech Stock Sell-off 2026: Why the Nasdaq is Dropping as Alphabet and AI Leaders Settle into a Bearish Reality

Click to comment

AI

Apple vs OpenAI Lawsuit: The Economic Story Behind the Headline

Published

3 hours ago

July 13, 2026

Abdul Rahman

Apple has sued OpenAI, alleging trade secret theft that the company says occurred “at every level” of its operations. Beyond the corporate drama, the case matters economically because it’s an early test of how courts will treat intellectual property disputes in an industry where enterprise customers are simultaneously investing hundreds of billions of dollars in AI infrastructure built on trust between a small number of vendors.

What actually happened

Apple filed suit against OpenAI, alleging a scheme of trade secret theft that the company characterized as occurring “at every level” of its operations, according to reporting picked up across financial and technology desks in July 2026 (CNBC). The filing lands at a moment when Apple’s own stock has been on an unusually strong run tied to the broader AI rally, illustrated in one widely circulated chart tracking how Apple shares “rode the AI rollercoaster to record highs” (CNBC).

Why this is an economics story, not just a legal one

Most coverage has treated this as a straightforward corporate dispute. The more consequential angle — and the one under-covered outside specialist legal and tech press — is what the case signals about vendor concentration risk in enterprise AI spending. Nvidia itself estimates that roughly 20% of its business comes from supporting frontier models built by OpenAI and Anthropic, according to TD Cowen estimates cited on CNBC’s markets desk, while Nvidia’s revenue from enterprise applications across other industries sits in the low-to-mid teens as a percentage of total revenue (CNBC).

That concentration matters because it illustrates how much of the current AI capital expenditure supercycle rests on a small number of foundation-model relationships. A high-profile IP dispute between two major players in that ecosystem — even one that doesn’t directly touch chip supply — raises the salience of vendor and IP risk for every enterprise now signing multi-year AI infrastructure contracts.

The broader AI-spending backdrop

The lawsuit lands during what markets are already describing as a shift in the AI investment narrative — from a race to build ever-larger models toward a race to build cheaper, more efficient systems (CNBC). That transition matters for the lawsuit’s economic stakes: if the industry is entering a phase where efficiency and proprietary techniques (rather than raw scale) become the primary competitive differentiator, trade-secret disputes like this one become more economically consequential, not less, because the contested IP is closer to the actual source of competitive advantage.

Connecting it to the inflation debate

There’s a second, more indirect economic link worth noting: strategists have flagged that ongoing AI infrastructure investment is, in the near term, contributing to inflationary pressure even if it proves disinflationary over the long run, according to market commentary tied to the same news cycle covering this lawsuit (CNBC) — a dynamic directly relevant to the Fed’s decision-making, covered in our Kevin Warsh Fed doctrine piece. Legal disruption to any major AI vendor relationship has the potential to affect the pace of that capex cycle, which in turn feeds back into the broader inflation and growth debate playing out across every market covered in this batch.

What businesses should take from this

For any organization with meaningful AI vendor dependency, the practical lesson isn’t about the specific legal merits of Apple’s claims — it’s a reminder to build contractual and architectural flexibility into AI vendor relationships now, before disputes of this scale become the norm rather than the exception. Concentration risk in a handful of foundation-model providers is no longer a theoretical concern; it’s playing out in real time in courtrooms as well as capital markets.

Discover more from The Economy

Subscribe to get the latest posts sent to your email.

AI

AI Capex Bubble 2026: The Hidden $662B Debt Nobody Reports

Published

2 days ago

July 11, 2026

Abdul Rahman

Every earnings season now brings a fresh wave of headlines about hyperscaler AI capital expenditure hitting a new record. The “big four” — Amazon, Microsoft, Alphabet, and Meta — are on track to spend roughly $725 billion combined in 2026, a 77% jump from the $410 billion deployed in 2025 (UnboxFuture). That number gets reported constantly. What almost nobody is reporting with the same prominence is a separate figure that may matter more: roughly $662 billion in data center lease commitments that hyperscalers have already signed but not yet begun — obligations that currently sit entirely off balance sheet.

Why the Off-Balance-Sheet Number Changes the Whole Picture

Under GAAP accounting rules governing when a lease “commences,” these signed-but-not-started commitments don’t appear in the capital expenditure figures analysts and investors typically scrutinize when assessing hyperscaler financial health. According to reporting citing Moody’s early-2026 analysis, this shadow liability is larger than the combined on-balance-sheet debt of the same companies (Anomaly Investments).

That detail matters enormously for one specific argument AI infrastructure bulls have relied on: the claim that this buildout is being conservatively self-funded from operating cash flow rather than risky leverage. Once the full picture of committed-but-unrecognized obligations is accounted for, that defense becomes much harder to sustain.

The Debt Is Already Showing Up, Not Just Theoretical

This isn’t a purely hypothetical concern about future liabilities. Big tech companies have already issued more than $100 billion of bonds in 2026 specifically to help fund AI capital expenditure, and investors have responded by demanding record levels of protection against potential defaults through credit default swaps — essentially insurance policies against bond default (IEEE ComSoc).

Individual company examples illustrate the shift toward leverage: Oracle issued an $18 billion bond specifically tied to its data center expansion; CoreWeave secured a $2.6 billion loan alongside a $1.75 billion bond package; and OpenAI and Oracle reportedly entered into a $100 billion vendor financing arrangement (Anomaly Investments). At Amazon specifically, capital expenditure over the trailing twelve months has reached $151 billion — a figure that now exceeds the company’s entire operating cash flow, pushing free cash flow into negative territory.

The Depreciation Assumption Almost No Coverage Questions

Here’s an angle genuinely underexplored across most financial media: the depreciation schedules hyperscalers use for AI hardware assume a five-to-six-year useful life. But given how rapidly GPU generations are turning over and how intensively AI workloads are pushing hardware utilization, critics argue the real economic life of this equipment is closer to two to three years. That gap between assumed and actual depreciation is estimated to understate true asset depletion by roughly $176 billion between 2026 and 2028 alone — a figure that grows as accelerating token consumption pushes hardware utilization beyond the assumptions built into current depreciation schedules (Anomaly Investments).

Layered on top of that is the energy cost curve: running the current roughly 30-gigawatt installed base of AI infrastructure costs approximately $27 billion annually today, but that figure is projected to climb to between $45 and $90 billion per year as capacity scales toward 2029 — and crucially, these are first charges against revenue, not optional or deferrable costs.

The Revenue Gap: Who’s Actually Paying for All This?

The most commonly cited justification for the capex surge is that the pure-play AI vendors — OpenAI, Anthropic, and others — represent a massive and rapidly growing revenue opportunity. The reality is more nuanced. OpenAI’s roughly $20 billion annualized revenue run rate, while genuinely impressive for a company with barely any consumer products three years ago, represents only about 3% of projected 2026 hyperscaler capex. Anthropic’s roughly $9 billion run rate, despite showing 9x year-over-year growth, occupies a similarly small share. The entire cohort of pure-play AI vendors combined — including Cohere, Mistral, Perplexity, and others — likely accounts for less than $35 billion in projected combined 2026 revenue against a hyperscaler capex figure exceeding $700 billion (Futurum Group).

That gap is the crux of the bubble debate: hyperscalers are betting the infrastructure will ultimately serve enterprise adoption and their own AI services broadly, not just third-party AI vendor revenue — but that bet requires enterprise AI monetization to arrive at a scale that, as of mid-2026, remains largely unproven outside of code generation and basic customer service automation.

The Skeptic’s Case, From Inside Goldman Sachs Itself

The most prominent voice of institutional skepticism doesn’t come from an outside critic — it comes from within Goldman Sachs itself. Jim Covello, the bank’s Head of Global Equity Research, has consistently argued the economics of the generative AI transition are fundamentally flawed, stating in mid-2026 that the industry has moved “further away” from justifying the scale of capital expenditure compared to two years prior (UnboxFuture). Covello has specifically flagged circular capital flows between cloud providers and AI startups — where hyperscalers invest in AI companies that then spend that same capital purchasing compute from those same hyperscalers — as a red flag reminiscent of vendor financing patterns seen in the dot-com era.

The valuation comparison to that era is explicit and increasingly common among strategists: US technology and AI equities carry EV/EBITDA multiples near 25x, close to historical extremes and above the telecom valuations that preceded the 2000 dot-com peak. More specifically, capex is currently expanding roughly 46 percentage points faster than revenue growth — a gap that exceeds the 32-point divergence observed during the 2001 telecom excess cycle (Allianz Research). Separately, Bank of America strategists have pointed out that AI stock concentration has reached levels matching prior bubble peaks, with the “AI Big 10” (Nvidia, Microsoft, Alphabet, Amazon, Meta, Apple, Tesla, Broadcom, Micron, and AMD) now making up 41% of the S&P 500 — comparable to the concentration of tech and telecom stocks during the actual dot-com bubble (Yahoo Finance).

The Bull Case Isn’t Naive Either

It would be inaccurate to frame this purely as informed skeptics versus blind enthusiasm. Goldman Sachs’ own broader research (distinct from Covello’s individual view) models roughly $7.6 trillion in cumulative AI capital expenditure between 2026 and 2031, built on the expectation that token consumption will increase 24-fold by 2030, driven largely by enterprise AI agents becoming embedded in production workflows rather than remaining experimental (Sesame Disk / Goldman commentary). Microsoft has disclosed an $80 billion backlog of Azure orders it currently cannot fulfill due to power constraints — genuine evidence that demand, at least for existing capacity, is outpacing even the current aggressive build-out pace (Futurum Group).

Leverage levels also remain more conservative than headlines suggest in absolute terms: the top five US capex providers reported a combined $385 billion in debt at the end of 2025, with leverage ratios still roughly 20% below the “high spender” cohort from the 2000 dot-com peak, according to Allianz Research analysis — meaning rising debt levels are a trend worth monitoring closely, not yet an acute crisis.

What Happens If the Bubble Skeptics Are Right

Historical infrastructure cycles offer a specific and somewhat counterintuitive lesson: the investors who fund the initial frenzied build-out phase rarely capture the long-term rewards. If the AI capex cycle follows the pattern of the 1998-2001 fiber optic buildout, hyperscalers may eventually be forced to write down the value of data centers and GPUs purchased at today’s prices and utilization assumptions. But that collapse in computing costs, paradoxically, could pave the way for a new generation of leaner, genuinely profitable software companies to build on top of the resulting cheap, overbuilt infrastructure — much as fiber-optic overbuild eventually enabled the 2000s streaming and cloud computing boom, even after the original telecom investors were wiped out.

What This Means for Investors and Businesses

For equity investors, the practical signal to watch isn’t the headline capex number — it’s the widening gap between capex growth and revenue growth, and whether that gap begins narrowing through 2027 as enterprise adoption either accelerates or disappoints. For businesses evaluating AI vendor relationships, the circular-financing pattern flagged by Covello is worth diligence: understanding whether an AI vendor’s revenue depends partly on capital originally supplied by the same hyperscaler providing its compute is a legitimate red flag for assessing that vendor’s underlying financial independence. For fixed-income investors, the rising credit default swap pricing on hyperscaler-linked debt is itself a market signal worth tracking as an early indicator of shifting sentiment, independent of equity price action.

The Bottom Line

The AI infrastructure buildout genuinely is the largest corporate capital expenditure cycle in recorded history, and it’s happening for real, defensible reasons tied to a genuine technology shift. But the debate over whether it constitutes a bubble isn’t really about whether AI technology is useful — it’s about whether the timing of returns can keep pace with public equity markets’ patience, and whether the $662 billion in off-balance-sheet lease commitments, aggressive depreciation assumptions, and circular vendor financing arrangements represent manageable financial engineering or the early architecture of a genuinely serious correction. Both cases have real evidence behind them. What’s clear is that the headline capex figure everyone quotes is no longer the most important number in this story.

Discover more from The Economy

Subscribe to get the latest posts sent to your email.

AI

AI Bubble Warning 2026: Why BIS, IMF and Bank of England Fear a Market Crash

Published

1 week ago

July 5, 2026

Abdul Rahman

Global financial regulators have moved from quiet skepticism to open warning, marking one of the most significant shifts in central-bank rhetoric since the aftermath of the 2008 crisis. The Bank for International Settlements (BIS), the International Monetary Fund (IMF), and the Bank of England have each flagged the risk that a correction in artificial-intelligence valuations could cascade through the global financial system, according to the BIS Annual Economic Report 2026 and reporting compiled by Wikipedia’s tracking of the unfolding episode.

From Confidence to Contagion Fear

The warnings did not emerge in a vacuum. In late June 2026, South Korea’s KOSPI index was forced into a trading halt after Samsung and SK Hynix shares each lost roughly 12% in a single morning, a shock that rippled into the Nasdaq, which fell 2.2% the same day. By the following week, Oracle had recorded its worst trading week since the dot-com crash, sliding 19%, after Apple raised product prices in response to soaring chip costs. The sell-off, detailed in Wikipedia’s account of the June 2026 rout, spread across global chip manufacturers before the BIS issued its formal caution on June 29.

Pablo Hernández de Cos, general manager of the BIS, framed the moment as one of “progress” colliding with “peril,” pointing to inflationary pressure, elevated public debt, and what the institution calls AI exuberance as compounding financial vulnerabilities.

Why This Cycle Looks Different — and Why It Doesn’t

Comparisons to the 1999–2000 dot-com bubble are now routine among Wall Street strategists. Deutsche Bank’s global economics team has described 2026 as resembling “1999 meets 1990,” according to Fortune’s coverage of the growing exuberance debate. JPMorgan’s chief executive Jamie Dimon has repeatedly used the phrase “irrational exuberance,” borrowed from former Fed chair Alan Greenspan, to describe dealmaking activity that he says is running “gung-ho.”

Yet analysts at Fidelity note a structural difference from 2000: hyperscalers are largely funding AI capital expenditure from earnings rather than debt, keeping the capex-to-free-cash-flow ratio below 1, compared with nearly 4 at the dot-com peak, based on Fidelity’s bubble-indicator research. That distinction matters for systemic risk, since debt-fueled busts tend to transmit further into the banking system than equity-only corrections.

The Systemic Transmission Risk

Oliver Wyman’s analysis of a potential AI-led market collapse estimates that an equity crash on the scale of the early 2000s could erase approximately $33 trillion in value — more than annual US GDP — a scenario that would compound if financing tied to data-center and digital-infrastructure debt turns out to be more opaque than banks currently report, according to Oliver Wyman’s assessment of financial-sector exposure. US equity market capitalization currently sits at close to twice GDP, a higher multiple than at the dot-com peak.

Prediction markets have already begun pricing the risk. Polymarket data cited by Tekedia shows the probability traders assign to an AI investment-frenzy collapse by the end of 2026 climbing to 26%, up sharply in recent months as valuations in chip and hyperscaler stocks stretched further.

What Regulators Are Asking Institutions to Do

The BIS is not calling for a halt to AI development. Instead, it is urging financial institutions to build greater transparency into AI-related financing, particularly the private-credit channels that now fund a large share of data-center buildouts, and to stress-test balance sheets against valuation drops of 30%, 40%, or even 50% in AI-exposed equities. The Bank of England has separately warned that investors have not been adequately cautioned about downside scenarios tied to companies such as OpenAI, whose valuation more than tripled between October 2024 and the following year.

For markets in the UK, US, Singapore, and East Asia’s chip-manufacturing hubs, the message from regulators is consistent: the innovation is real, but the financing structure underneath it has not been fully stress-tested against a reversal in sentiment.