🦔Moody’s says AI borrowing has become a major source of growth for private credit, and it sees a problem coming. A wave of these loans comes due between 2027 and 2029, and weaker borrowers will have to refinance at higher rates. About 17% of the 607 companies Moody’s tracks already pay part of their interest with more debt instead of cash. My Take With payment-in-kind, a company that can’t spare cash for its interest bill gets the lender to add the interest to the loan instead. That buys time. The debt grows every quarter while everyone waits for business to improve before the loan comes due. With about one in six borrowers already on it before the big maturity wave even starts, I don’t love the odds for 2028. Moody’s also used the word “circular,” and rating agencies usually show up late to that kind of call. Nvidia alone has put up to $164.5 billion behind its own customers’ data centers, so when one piece of this chain breaks, the lenders and the chip suppliers take the hit together. Expect private credit funds to push maturities out and keep these loans marked near full value as long as they can. If you own one of the private credit funds that cap withdrawals, the value on your statement depends on those marks. Hedgie🤗

Credit spreads in the tech sector now exceed those of other sectors, as the tech industry shifts from being a cash generator to a cash incinerator.

THE AI TRADE IS FACING A CRITICAL TEST: The world's four major memory chipmakers, SK Hynix, Samsung, Kioxia and CXMT, report earnings this week, This puts a near-$1 trillion industry and more than $2 trillion in combined market value in focus. The industry is benefiting from semiconductor shortages and soaring prices, with the supply crunch expected to persist, yet rising financing costs and concerns over future returns are increasingly challenging the outlook. Samsung and SK Hynix shares are already down roughly -50% from their June peaks as investors increasingly question whether AI demand will generate sufficient returns to justify the enormous cost of data centers and computing infrastructure. The two Korean giants together account for ~60% of global DRAM revenue and over 80% of high-bandwidth memory revenue, while their combined operating profit is estimated to reach 1,571 trillion won ($1.2 trillion) across 2026 and 2027. This week's results could determine whether the AI chip stock selloff has further to run.

the biggest AI spenders have sold $148bn of stock so far in 2026. add meta's $35bn launched on 26 october, its planned $60bn and the roughly $100bn anthropic is reported to be raising, and AI is asking the stock market for about $340bn inside a year, with the fed hiking. the bond market already showed its limit. amazon, alphabet, meta and oracle sold about $194bn of bonds by 7 july against $108bn in all of 2025, and order books fell from nearly five times covered to under two. equity is the window still open, which is why they are all going through it at once.

JPMAM: If rates fall 1%, you'll earn 13% on the 10yr Treasury... if rates rise 1%, you'll lose 2% $TNX

GPU RENTAL PRICE HIKES BY CLOUD PROVIDERS: • AWS: Q3 increase of 20% • Nebius: cumulative 50% increase • CoreWeave: 25% across all SKUs $AMZN $MSFT $GOOGL $CRWV $NBIS $IREN $DOCN

*ANTHROPIC TO RELEASE S-1 AT 9AM EST: FOX BUSINESS

$META announced plans to raise approximately $35 billion through two offerings: > $22.5B in Class A common stock through forward sale agreements. > $12.5B in depositary shares representing mandatory convertible preferred stock. Interestingly, NVIDIA, AMD and Broadcom have expressed nonbinding interest in purchasing up to $5 billion of Meta shares combined. Meta is also planning a potential $60 billion offering program for its new nonvoting Class C shares starting in Q1 2027. Together, the plans could provide approximately $95 billion in equity financing capacity as Meta prepares to spend $135B to $145B on capital expenditures in 2026.

J.P. Morgan: global server market is estimated to expand to $506 billion in 2026 from $284 billion in 2025, implying +78% Y/Y growth, while the long-term CAGR from 2026 through 2030 is expected to track to +31%. $DELL $HPE $SMCI

WSJ: "Wall Street has rested much of its optimism about the stock market on a surge in corporate earnings. The catch is that the profit growth is being driven by a handful of AI giants. Analysts expect companies in the S&P 500 to report a 30% jump in third-quarter earnings from a year ago, boosting investor confidence that the good times can keep rolling so long as businesses keep printing money. But just two companies—Nvidia and Micron—account for 34% of that growth. And only 10 companies—including Broadcom and Meta—account for roughly two-thirds of the profit growth." Especially as investor concern about debt levels and AI CAPEX levels intensifies, it's essential to recognize the role circular spending is playing in propping up profit growth. AI is not playing a role in boosting profitability outside of tech where competition for limited resources is inflating the cost of compute. To quote Torsten Slok: "Despite massive investments and widespread hype around artificial intelligence, the technology has delivered zero financial payoff to corporate America outside of tech. Revenue per worker is rising for the Mag 7 but is flat for the S&P 493. Profit margins are rising for the Mag 7 but are flat for the S&P 493. And there are still no signs of AI boosting productivity growth." As I warned in my September report on "The AI Trade" (sageroadresearch.com/collect…), the impact of circular spending will recede even if AI CAPEX doesn't contract. To quote the report: "Even before considering the viability of AI ROI, comps are a significant concern for US equities over the next 12 months. UBS has estimated that hyperscaler AI CAPEX will increase roughly 75% this year, 25% next year, and only 6% in 2028. That rate-of-change slowdown will certainly hit the semiconductor industry the hardest and that’s undoubtedly contributed to Nvidia’s stock performance this year, only slightly outperforming the S&P 500 despite blockbuster earnings growth (the company is expected to account for a staggering 18% of the S&P 500’s total YoY earnings growth in 2026). However, we also believe market participants have neglected the impact circular spending is having on hyperscaler growth and the trickle-down impact to companies across sectors. As the rate of increase decreases, we expect earnings growth to broadly slow. And that’s before considering how everything we’ve laid out in this report could lead to spending that falls below estimates." I do expect a pullback in CAPEX as rates stay higher for longer and AI ROI continues to fall short of investor expectations. The market is punishing plummeting FCF regardless of earnings growth--see Bloom Energy post earnings this week. In September, I told my clients to reduce, if not eliminate exposure to AI picks and shovels. I stand by that today. Message me if you'd like to learn more about Sage Road Research and our investment track record.

Lately, the 10Y is *seemingly* being driven by AI CapEx expectations. SOXX is the proxy. 87% correlation over training 9-weeks. P-value = 0.003. The deepest and most liquid market in the World is *seemingly* confirming what the equities are screaming. CapEx will fall short of prior expectations. Correlation ≠ Causation

APOLLO: AI HAS DELIVERED “ZERO PAYOFF” OUTSIDE BIG TECH Apollo Chief Economist Torsten Slok warns that massive AI spending has yet to deliver measurable productivity gains across corporate America. Revenue per worker and profit margins are rising for the Magnificent Seven but remain flat for the S&P 493. Meanwhile, Gartner estimates $3.5 trillion in AI infrastructure spending could face impairment risk by 2030. The key takeaway: AI investment is booming, but returns outside Big Tech remain elusive.

High yield officially negative total return for the year (-0.09%). Overall returns across all ratings look terrible. CCCs especially at -5.1% YTD and spreads over 1,100bps now. Lots of idiosyncratic stuff in there so would't read too much into it though.

Depreciation and amortization expenses for the five hyperscalers are set to reach $590 billion in 2029. This is likely to exceed their combined operating profits in 2026.

Just 2 years ago, Bessent mocked Yellen because she was using short-term bills to finance the government. Well, he's only accelerated that practice & a new IMF paper shows how dangerous that can be. America has relied heavily on short-term borrowing before, particularly in the 70s, 80s and early 2000s. The difference? Government debt was substantially lower relative to GDP. Today, with debt at historically elevated levels, greater reliance on short-term borrowing leaves the Treasury increasingly exposed to interest rate changes. If rates stay higher for longer, refinancing costs could rise rapidly, putting further pressure on deficits. The irony? The Fed's fight against inflation could make America's debt problem even harder to manage.

GPU prices higher and token prices lower means the crunch is compressing. That means lower profit margins on the GPU rental to token production leg of the trade. Of course the profit margin on buying GPUs and renting them out is higher than ever (not exactly an indication of a slowdown in inference demand).

OpenAI & Anthropic set their prices based on venture capitalists, who care about founder vision + tech talent. Nvidia’s price is set by Wall Street investors, who care about margins + profits. We don't have an AI bubble. We have a startup bubble. Look below; the difference is quite something.