Executive Summary:
The AI Reality Check & Debt Loop: The technology sector is enduring a vicious correction – heavily concentrated in chip stocks – as investors scrutinize the massive, increasingly debt-funded AI CapEx cycle and face growing local political resistance to new data center construction.
Actions Have Consequences: The midterm "Blue Wave" has ushered in a divided government, neutering a pro-Wall Street legislative agenda and signaling a potential long-term policy shift toward addressing historic wealth inequality at the expense of corporate margins.
Navigating the Volatility: As the AI narrative shifts from soaring demand to precarious financing, investors must accept heightened two-way volatility as the price of admission, favoring strong balance sheets and unemotional discipline over momentum and narratives.
Full Commentary:
The bombshell news to start the week is the revelation that Nvidia’s CEO Jensen Huang acknowledged being the target of a foreign espionage effort. Huang was involved in a romantic relationship with an executive suspected of working for the Chinese Intelligence A gency, prompting the board to launch a probe into the issue. The share price is trading 4-7% lower in today’s trading session, compounding the severe pressure already mounting from recent doubts over the sustainability of the AI boom. This matters deeply because Jensen, much like Elon Musk, is the defining face of one of the largest companies by market cap in the world.
Simultaneously, whispers are emerging from the Middle East suggesting that ongoing diplomatic talks are making tangible progress toward a resolution of the U.S./Iran conflict. This would be a significant development to relieve pain at the pump for consumers, who made the Iran War and its impact on affordability a central issue for voters who ushered in the recent ‘Blue Wave’ at the midterm elections. As of this writing, Brent crude is down an additional -4.1% to $94 per barrel, marking a significant drop from the $117 level it reached in late October.
Looking beyond tomorrow's CPI report, and what looks like concerted interest in finding an offramp for this ‘skirmish’ in the Middle East, it appears inflation prints will start to roll over as we slip into 2027. Oil prices are slipping into a lower range, unit labor costs are declining, and market-based inflation expectations have come down considerably. Copper quietly reversed to a four-month low, dropping nearly 3% for the week. All the while, real 10-year yields have declined by 32 basis points, falling from 2.95 to 2.63%. Even with this drop, they remain higher than at any point over the last two years, excluding the past two months, safely above the old high-end band of 2.40 percent.
The Tech Wreck and the AI Reality Check
It has been a rough stretch beneath the surface of the equity markets for several months, but last week the carnage finally carried over into the major indices. The S&P 500 was battered by -2.7%, the Dow slipped -0.94%, and the Nasdaq was pummeled by -4.91%. If you were long risk assets with exposure to the AI renaissance that has been gaining momentum since ChatGPT launched in October 2022, the week was brutal.
The fallout from the election results – where Democrats exceeded even the most optimistic expectations – put a firm bid under hyperscalers like Amazon (+7.9%), Alphabet (+2.1%), and Microsoft (+1.7%). The market quickly determined that their existing data center fleets are now inherently more valuable, given that the timing and scale of future capacity are suddenly in question.
Conversely, the damage in the tech sector was heavily concentrated in the chip stocks that rely on future infrastructure growth. Nvidia fell 8.8%, Broadcom dropped 9.9%, and Micron plunged 22.0%. A slower, more heavily regulated AI rollout explains both sides of this dynamic: it keeps CapEx in check for the hyperscalers while inflating the moat around their current operations, but it raises severe concerns about semiconductor order rates and backlogs moving forward.
Outside of Apple – which made a concerted effort to stay out of the AI arms race – and two of the best-capitalized hyperscalers (Microsoft and Amazon), the decline in this space has been breathtaking. The average drawdown from 52-week highs approaches 35%, with the median sitting just under 31%. Keep in mind that these are not low-quality micro-caps; they are long-standing, time-tested, and fundamentally strong operators. These drawdowns illustrate just how nasty a correction some areas of the capital markets have had to endure over the last several months, especially when compared to an S&P 500 that remained a few percentage points shy of even breaching the 10% correction threshold.
That being said, we are nearing, or may have already reached, a short-term capitulative extreme. Sentiment and positioning have undergone a thorough cleansing, and prices in the AI names have retraced virtually all of their post-Iran war runup. The Nasdaq 100 has pulled back to a key technical level, re-testing its base trendline that connects the Liberation Day low of April 2025 with the March 2026 low. How markets react off this important technical level will be key for the rest of the year.
Fascinatingly, capital rotating out of tech is seeking refuge in consumer staples, pushing the sector's valuation premium against the S&P 500 to its highest level in over a decade. The spread sits at roughly 6 P/E points versus a long-run average of 1. Jim Chanos posted the following statement on X that was just too on point not to share:
“I can’t shed the feeling that selling high-beta tech stocks that are already down 40-70% to buy low-growth consumer names at 30-100x earnings… might not be a smart move, either.”
I couldn’t agree more with Jim, it makes little fundamental sense for toothpaste and cereal to trade at twice the multiple of the chips running global data centers.
The AI Debt Loop
The central question the market is now wrestling with is how to continue funding an enormous CapEx commitment – specifically, where the capital comes from and who gets crowded out in the process – and how the Return on Investment (ROI) profile fundamentally shifts in the wake of the 'Blue tsunami.' When the pace of investment stops accelerating, profit growth expectations inevitably begin to sink. With funding costs marching higher, the entire AI ecosystem is discovering that swimming against the credit cycle is becoming a prohibitively expensive endeavor.
This equity swoon is not an indictment of the technology itself, but a repricing of prior assumptions and exuberant investor behavior. As was the case with the Internet, investors who stay at the dance too long in companies that become casualties of the technology will get their capital wiped out. Yahoo, Alta Vista, and AOL are gone, but companies with solid business models and diversified product lines like Microsoft, Cisco, Amazon, and Apple survived the 2000-2002 Tech wreck.
The AI buildout is increasingly running on borrowed money. According to Goldman's credit desk, 36% of 2026 CapEx is debt-financed – up from 33% – reflecting front-loaded issuance and negative free cash flow. Hyperscaler debt issuance is projected to peak near $470 billion in 2028, with capex marching toward $1.57 trillion by 2030. The four main hyperscalers have increased their capex 6x to 9x since 2023, while Apple’s capex has actually declined by about a third.
Economists Jared Bernstein and Ryan Cummings highlighted this danger in a very insightful Op-Ed piece in today’s NYT (Tech Stocks Had a Terrible Day. What’s Coming Could Be Worse ), noting that AI infrastructure companies are no longer just spending their own cash. In the first eight months of 2026, $445 billion worth of AI-related debt was issued, equating to 1.4% of GDP. Including opaque off-balance-sheet deals, total AI infrastructure debt is roughly 7% of GDP, approaching the subprime mortgage lending levels that helped trigger the financial crisis.
Furthermore, the Bank for International Settlements reports that almost 30% of AI firms’ investment deals are in other AI firms, and over half of the investments in AI companies come from the ecosystem itself. This includes chipmakers funding their customers and cloud giants owning stakes in AI labs, creating a cross-investment loop that turns very fast when the cycle shifts.
The Midterm Referendum on Data Centers
The midterms were as much a referendum on the AI data center construction boom as anything else. Across six states, there were 57 ballot measures concerning data centers, and the industry lost 46 of them. Recent polling shows 71% of Americans oppose building local data centers, compared to just 53% for nuclear plants.
States taken by Democrat governors who pledged moratoria on new AI data center construction represent 45% of new planned capacity. Ohio alone saw voters pass 21 local ballot measures aimed at restricting development. About 40% of pending data center capacity in the U.S. is planned in states where newly elected leaders have pledged pauses or cancellations, putting nearly one-third of compute power projected for 2028 at risk.
This is a major shift for local government officials who, under pressure from Washington, had been offering huge tax breaks and other incentives to mlure data center developers to their communities. Now, the tables have turned and the balance of power has shifted to environmentally sensitive Democratic leaders—and this new group is going to be pushing data centers to pay taxes, create their own power plants, pay for their use of electricity and water, and contribute to local community resources like education funds.
It must be emphasized that the recent market weakness was taking hold before the midterms, but what this new political environment does is complicate the landscape even more. The pace of growth in spending on chips and data centers was already exerting a strain on the capital markets, a key factor in sending borrowing costs higher across the board and creating backlogs. And what the sharp compression in P/E multiples has been signaling has nothing to do with demand or current profitability growth, but rather marked-down future ROI assumptions with each dollar spent and borrowed. Even before the political shift, it has been the market itself and the intense competition for capital that placed the constraint on these breakneck CapEx plans.
While the past week’s turmoil is probably not the last whipsaw we are going to see as the cycle plays out, the main message is not one of Armageddon, but rather the market doing its job ahead of the political class in placing constraints. What investors are going to have to do now is decide whether to bail entirely or sharpen their pencils to discriminate between the winners and losers. The more diversified the business, the better; the more of a pure-play you are in AI, the more trouble you’re likely going to face.
That has a wide range of implications for the different players. Companies like Amazon and Alphabet that have solid core businesses and easy access to capital markets could be relatively insulated – they’ll just slow down their spending on AI CapEx. Companies like Oracle and CoreWeave, which are more highly leveraged and more dependent on continued growth in AI spending, look much more tenuous. Apple, meanwhile, seems to be playing from a position of tremendous strength.
But it must be stressed that what last week’s market action did was expose how much of the AI boom rests on a fragile web of cross-investments. The BIS reports that almost 30% of AI firms’ investment deals are in other AI firms, and more than half the investments being made in AI companies come from other AI companies—chipmakers funding their customers, cloud giants owning stakes in the AI labs that rent their capacity, and leases and guarantees that sit largely off the balance sheet. When that loop turns, it turns fast. When I say that the hyperscalers are winners, that is in a relative sense only. This past week was a week when the market got the political clarity it wanted, then discovered a new problem: the election didn’t remove uncertainty. It just moved it. Political risk went down, perhaps, but AI risk went up.
Finally, how much of this new dent in the GDP growth outlook is going to factor into the Fed’s thinking? A good chunk of Kevin Warsh’s growth and inflation prognosis was hitched to the AI spending boom. That boom hasn’t necessarily ended, but the shape and movement in the demand curve are surely now less pro-growth and less inflationary.
Gridlock and Wealth Inequality
While the initial investor reaction to the midterm election results was surprisingly positive, it cannot be debated that the typical equity investor lost a very good friend: one-party Republican rule. Donald Trump still controls the executive branch but now has zero influence on the legislative process. The market-bolstering tax cuts of 2018 and this year’s One Big Beautiful Bill would never have lived to see the light of day if the Democrats had control of one of the chambers of Congress. Now they own both. Trump campaigned on the message that in his second term he was going to focus on ‘Main St.’ over ‘Wall St.’, but based on every consumer sentiment poll heading into the election—including his all-time low poll readings relating to the economy – the median voter determined his administration wasn’t delivering on that message and went in another direction. Well, now it’s the Democrats' turn, and they don’t have Wall Street as their prime constituency and do not measure success by how the S&P 500 performs.
One last thought on the landslide Democratic victory. Without question, the Iran war played a major role in how people voted, but I’d argue it is the impact the Iran war had at home – further exacerbating pressure on household pocketbooks – that was the straw that broke the camel's back for the GOP. So far in Trump 2.0 and under full GOP control, wealth and income inequalities have only grown larger, as is clearly demonstrated in the charts below from David Rosenberg
Capitalists are going to have to think about what the future brings if the Democrats attempt to mean-revert the insane gaps in these charts—not that any attempt to do so is wrong. After all, voters have been lurching from one politician to the next who tells them the disparity in these charts isn’t working for the masses. If and when elected officials actually deliver on such promises, I don't expect it to be friendly to the stock market. I think it is highly probable that post-2028, we will see tax reform designed to preserve social stability, but it will come at the expense of corporate income to support work-based personal income, which, as you can see, has nosedived to its lowest level on record.
Navigating the Volatility
So, my advice to investors is to respect and recognize all the statistics being thrown around about constructive seasonality, how the equity market performs following mid-terms, and the third year of a Presidential cycle – which are all strong, positive historical tailwinds. But keep them in context with other variables coinciding with the investment cycle, capital demands, monetary policy, and fiscal room. Capital markets are never as simple as just one thing, no matter how much we’d like them to be. The historical record is important and useful, but we also don’t know yet what the catalyst is going to be that breaks markets out of their current funk. After 2018, the Fed was busy cutting rates; in 2020, the government went wild with fiscal stimulus, and the Fed cut rates to zero and opened up its balance sheet; in 2022 we had ChatGPT; and in 2025 we had Liberation Day followed by tariff reversals galore (oh yes, and investment deals worth trillions). Stay vigilant, disciplined, and adapt your plan along the way.
As much as we’d all like to compound our portfolios at 10% per annum with no volatility, that’s not how equity markets work. Historically, equity markets have compounded at a superior rate to bonds and other asset classes because they have a higher risk profile. That risk goes in both directions – upside and downside. So as uncomfortable and as queasy as these periods make you feel, don’t lose sight of the reality that a higher level of volatility is the price of admission.
I know, no one wants to hear this, just as I don’t want to write it. It feels like I’m lecturing my 12-year-old daughter. What is important is having the confidence, trust, and unemotional fortitude to trust your allocation, process, and discipline (assuming you have one; if not, get on it quickly) to get you through these periods. No, that isn’t an endorsement to do nothing and stand your ground if you are improperly positioned for a regime change that is underway. It’s a suggestion to be measured and thoughtful with portfolio pivots that make sense. Don’t allow your emotions to get the best of you. Understand that you are not capable of catching every twist and turn in an equity market, even if the talking heads on bubblevision make you feel as if they have when the VIX crosses above 25 on its way to higher levels.
Bottom line: move forward, but do so with one eye on the long-term plan while keeping the other on the short-term tactical adjustments that can and should be made to smooth the path to those long-term objectives. My guess is that none of the long-term planning you’ve done—if it included equities as a tool in the asset class repertoire—assumed zero downside volatility along the way.
Expect volatility to stay high. The AI narrative is shifting from demand to financing, which is a much harder story to sell. Favor balance sheets over narratives—look for companies that fund growth from their own cash flow rather than relying on customers or credit markets. The cloud giants have the most cushion, while debt-funded builders have the least. Finally, do not expect a quick rescue from the Fed; they are still leaning tighter in a market that remains over-owned and priced for aggressive growth.
The articles and opinions in "Capital Market Musings and Commentary" are for general information only, and not intended to provide specific investment advice. Performance, dividends and other figures have been obtained from sources believed reliable but have not been audited and cannot be guaranteed. Past performance does not ensure future results. Investing inherently contains risk including loss of principle. Advisory services offered through Casilio Leitch Investments, an SEC registered investment advisor. Copyright © 2026 Casilio Leitch Investments. All Rights Reserved.