Speaker · Updated
Patrick Boyle: latest market views
Patrick Boyle On Finance
Summary
Patrick Boyle is an Irish hedge fund manager, university finance professor and former investment banker who hosts the Patrick Boyle On Finance YouTube channel and podcast on quantitative finance, markets and financial history.
Chorus has 20 views from Patrick Boyle across 2 episodes on Patrick Boyle since October 25, 2026. Their latest calls: bearish on Anthropic, CoreWeave, Oracle and QTS Realty Trust.
Where they stand
Their latest bullish or bearish call on each company.
What they talk about most
The latest views
Newest first. Every quote is checked against the episode's transcript.
AnthropicNeutral
Anthropic's $80B first-half net loss is mostly an accounting artifact from valuation-linked instruments, not cash burn.
Of the $80.7B net loss for the first half, more than $72B comes from revaluing financing deals, mostly with Google and Amazon, that are treated as liabilities and convert into Anthropic shares. When Anthropic's valuation rose about $80B, those investors booked matching gains and Anthropic booked a matching loss, so no cash was burned. The loss grows as the business succeeds and will stop once the instruments convert. Anthropic strips this out of its adjusted profit of about $2.7B, while Amazon and Alphabet count the same swings as real gains. Boyle's actual operating loss was just under $700M for the half, a sharp improvement on about $2B a year earlier.
“but if you actually read the filing, more than $72,000,000,000 of that loss is an accounting illusion. It comes from the change in the fair value of some financing deals the company had done, mostly with Google and Amazon,”at 8:22
AnthropicBullish
Anthropic's economics are real: a Q2 operating profit and positive operating cash flow, even counting training costs.
Boyle says the numbers came in much better than the bears expected. Anthropic turned a genuine operating profit in Q2, and its computing spend fell from 251% of sales early last year to 48% in Q2. It generated almost $4B of positive operating cash flow without stripping model training costs out of the numbers. He calls the numbers real and good, but notes the financial picture stops at June 30, so it is backward-looking.
“Because the numbers came in much better than the bears expected, they turned a genuine operating profit in the second quarter. ... The numbers are real and they are good.”at 11:23
AnthropicBearish
Anthropic's Q3 sequential growth reportedly halved to about 45% amid price cuts, fading token-maxing and cheaper rivals.
Per private figures shared with investors (not in the SEC filing), sequential revenue growth fell from 147% in Q2 to about 45% in Q3. Boyle says 45% quarterly is still extraordinary, about 250% annualised, but a sudden halving is the kind of thing that makes prospective investors nervous. He attributes the first-half surge partly to token maxing, which Eisman argued had faded. Corporate clients then routed cheaper work to Chinese open-weight models. Unconfirmed reports say Anthropic cut prices sharply in October, the filing confirms cuts of 25-40% in the cost of a typical task, and Google priced Gemini to match Claude. He calls the price war good for customers but badly timed before an IPO.
“According to figures Anthropic shared privately with investors and which are not in the public filing, its sequential revenue growth in the third quarter fell from 147% to around 45. Now, a 45% quarterly growth rate is still extraordinary... but a sudden halving of your growth rate is the sort of thing that makes prospective investors nervous.”at 14:57
AnthropicBearish
Anthropic's founder-controlled, trust-based governance leaves public shareholders with weak rights and little real check.
Boyle describes a layered structure. Anthropic is a public benefit corporation, so the board can weigh its mission against shareholder returns, and shareholders could not sue over a large safety spend. A Long-Term Benefit Trust with five seats, only four filled (including Ben Bernanke), nominates four of seven directors. But the seven co-founders hold a fixed 50.1% of the voting power, which is undiluted, elect three directors, and hold two of three nominating-committee seats, so they can block the trust's swing nominee. Deadlocks go to an employee stock vote. Amazon and Google, the largest investors, get non-voting shares. Boyle says shareholders used to demand a discount for reduced rights, but now the market does not mind.
“Anthropic's seven co founders jointly hold a special class of shares that gives them a fixed 50.1% of the voting power.”at 19:57
AnthropicBearish
Anthropic's $800B+ compute commitments make it a heavy industrial business, so today's operating profit may peak.
Boyle says the prospectus discloses new computing-power agreements since the end of 2025 with minimum obligations of $417B, including $110B take-or-pay with Amazon, about $161B of chip leases with Broadcom, and a SpaceX deal of up to $120.5B (about $3.1B a month once online). With Google, Microsoft and neoclouds, total future commitments exceed $800B. These sit in the notes as purchase commitments and lease obligations rather than traditional debt. Boyle argues the asset-light, rented-compute era is ending. Anthropic is becoming a heavy industrial business pouring hundreds of billions into concrete, cooling and chips that go obsolete, so the Q2 operating profit may be about as good as it gets for a long time.
“Anthropic has committed to over $800,000,000,000 in future spending. $800,000,000,000. That is more than the defense budgets of most continents.”at 26:52
AnthropicBearish
Amazon and Google are Anthropic's investors, landlords, sales channels and rival model makers, a risky dependence.
Boyle says Anthropic distributes its models through its own biggest competitors. Amazon and Google are at once its investors, its landlords, the cloud platforms it sells through and the makers of two rival AI models. They take a cut of sales and could in principle stop carrying Anthropic whenever they liked, which the filing admits. Anthropic's 'Anthropic Direct' strategy, building down the stack to own chips and data centres and up towards customers, is ambitious but staggeringly expensive, because buying the buildings and chips means paying to maintain and replace them as they go obsolete.
“Amazon and Google are at the same time Anthropic's investors, its landlords, the cloud platforms it sells through and the makers of two rival AI models.”at 27:37
AnthropicBearish
Governments are unlikely to leave Anthropic's powerful frontier AI in private hands; expropriation risk is real.
The filing warns of heightened scrutiny, political backlash or disproportionate regulatory action over its public AI-safety positions. Boyle thinks advocacy is a red herring. If the technology is powerful enough to transform the economy and pose catastrophic risks, governments will not leave it in private hands regardless of how quiet the company is. He cites the Pentagon designating Anthropic a supply-chain risk, which a federal court found retaliatory before Anthropic lost on appeal last month, and export controls that pulled models off the market. Government contracts are only about 1% of revenue, so the direct hit is small, but the risk is who ultimately controls the technology, and expropriation of the whole business 'wouldn't be that shocking today'.
“The thing pulling the government in is the power of the technology. If you take this prospectus at face value... then you have just described the kind of capability that no government anywhere lets sit entirely in private hands.”at 30:51
AI deals are circular: Amazon and Google fund Anthropic, which buys their compute, so valuations are entangled.
Boyle says valuing AI stocks is nearly impossible because the industry passes the same money back and forth. Amazon and Google invest in Anthropic, Anthropic buys its compute from Amazon and Google, and it pays SpaceX billions a month for capacity still being built. Quoting de Moderen, to value Microsoft you now have to value OpenAI, and to value Google you have to value Anthropic. The technology is clearly real, but the sector is tied together in one very expensive knot, which makes its valuations hard to judge.
“this whole industry is passing the same money back and forth. ... The entire sector is tied together in one very expensive knot.”at 35:37
AnthropicBearish
Anthropic's roughly $2 trillion IPO valuation is impossible to justify on any AI multiple; the price is a guess.
Boyle applied Scott McNealy's payback test to a $2T Anthropic valuation in September and found it impossible to justify. On any AI company multiple there is a near mathematical impossibility of a reasonable payback period, and the filing does not change that verdict. He calls the price a guess. The talked-about valuation drifted from $2T to $1.5-1.75T within weeks, insiders are staggering lockups so they can sell soon after the IPO, and Meta and Alphabet are raising about $95B and $90B.
“Back in September, I ran Scott McNealy's old test on an anthropic $2,000,000,000,000 value, and it was impossible to justify. On any of these AI company multiples, there is a near mathematical impossibility of a reasonable payback period.”at 36:46
AI infrastructure & data centers
Data center water concerns are overblown; farm irrigation uses far more water than data centers.
Boyle says the water argument against data centers mostly falls apart when examined. Citing The Economist, he says a mid-sized data center uses about as much water as two golf courses. Crop irrigation across the country uses 74 times as much water as data centers. He suggests that if water were the real concern, protesters would also target country clubs. The point is that data centers are a small share of water demand.
“a mid sized data center uses about as much water as two golf courses do and across the country crop irrigation uses 74 times as much water as data centers do”at 3:15
Data centers aren't paying their own grid costs; ratepayers absorb future PJM capacity charges.
Boyle reports that EPRI found data centers built from 2019 to 2024 lowered power costs by about 6% in the typical state, but only because they used spare grid capacity that covered fixed costs. That spare capacity is now largely used up. The PJM Independent Market Monitor estimates about $6.3 billion of future capacity costs come from planned data centers, and he says the data centers are not paying those costs themselves. Boyle adds that if expected demand never arrives, utilities recover fixed costs from fewer customers, so the remaining customers pay for transmission lines. He treats both the earlier benefit and the later burden as true.
“...the Independent Market Monitor estimates that about $6,300,000,000 of future capacity costs come from planned data center costs”at 5:56
xAIBearish
xAI's portable turbine fleet exploits a permit loophole, imposing noise and smog on South Haven residents.
Boyle describes xAI bypassing normal power plant permitting by running gas turbines across the state line in Mississippi to supply its Memphis data center. Mississippi lets portable equipment run for up to a year without an air permit, so the turbines count as traffic. The company said it ran 27 turbines in January, and regulators counted 69 by July. An acoustics firm found low-frequency noise could be four to 16 times louder than background. The NAACP lawsuit claims the fleet could emit about 26 times the nitrogen oxides of the nearby TVA gas plant. Boyle concludes that the operation is unpermitted, even if not illegal, and that the community bears the costs.
“Now, normally building a power plant that rivals the largest in the state involves things like air permits and public hearings, but Mississippi allows portable equipment to run for up to a year without an air permit.”at 12:42
AI chips will become obsolete far faster than the buildings housing them, mismatching long-term loan collateral.
Boyle argues that much of the debt financing AI data centers is secured against the computer chips. Nvidia keeps releasing faster chips, so the chips inside will be obsolete long before the buildings, which last decades. Lenders are therefore lending long-term money against an asset that depreciates quickly. Boyle cites Carlos Mendez, who compared it to financing a fax machine just before email was invented. He presents this as a structural weakness in the financing.
“A lot of this debt is secured against the computer chips themselves, and Nvidia keeps releasing better and faster ones. The buildings will last for decades, but it is pretty much guaranteed that the chips inside them will be obsolete long before that.”at 16:24
QTS Realty TrustBearish
QTS's data center bonds repriced from 5.7% to 7.2% as lease-expiry refinancing risk surfaced.
QTS issued $4.6 billion of bonds in April at a 5.7% yield, mainly because Microsoft was the expected tenant. Investors then noticed the bonds would not be fully repaid before Microsoft's initial lease ends. That leaves QTS needing a new tenant for a very large building if Microsoft leaves. When QTS returned to the market in August, investors demanded 7.2%, about what junk-free top-rated companies pay. Boyle presents the rate increase as the market pricing in that refinancing and tenant risk.
“mostly because Microsoft was going to be the tenant, but then investors read the terms and conditions more carefully and realized that the bonds wouldn't be fully paid off before Microsoft's initial lease runs out”at 17:11
CoreWeaveBearish
CoreWeave's chip-backed high-yield debt looks dangerously risky, with insurance prices signalling default fears.
Boyle argues CoreWeave's data centers rely heavily on high-yield debt, much of it backed by its own chips, which become obsolete quickly. He cites the credit default swap market: insuring $10 million of its debt for five years costs about $1.95 million upfront plus $125,000 a quarter, roughly 11% a year. Bloomberg reports the swaps market prices a more than 60% chance of default within five years. He dismisses the co-founder's claim that insurance capital is unlimited, noting that insurers are reluctant to take on AI data center risk unless paid a large sum. Boyle regards the company's plan to win investment-grade ratings as an attempt to attract a different pool of buyers.
“Things are even worse at CoreWeave, the specialist cloud company I mentioned earlier, which relies heavily on high yield debt to pay for its data centers, a lot of it backed by its chips.”at 17:45
OracleBearish
Oracle's default insurance cost rose roughly tenfold, signalling rising credit risk from its AI buildout.
Boyle cites Matt Levine's observation that five-year credit default swaps on Oracle cost about 0.33% a year five years ago and about 3.3% today. He attributes the rise to Oracle being strapped into the AI buildout. Boyle presents this as evidence that the market now sees Oracle's credit risk as substantially higher because of its heavy AI spending. The argument rests on the change in CDS pricing rather than on Oracle's own financials.
“Five years ago, Levine points out you could buy five year CDS on Oracle for about 0.33% a year. Today, with Oracle strapped into the AI build out, the same insurance costs about 3.3%.”at 20:45
AI infrastructure & data centers
Data center costs are being pushed onto neighbours, ratepayers and pension savers, not the builders.
Boyle argues the public, not the tech companies, carries the cost of AI data centers. Residents face noise and lost property values, while a few towns and counties receive payments or settlements that are small against the scale of the projects. Pension funds, which he says are often his own, hold the shares and debt of the builders, so savers finance the buildout while also paying through power bills. If the boom fails, grid costs and losses fall on the same households. He concludes the neighbours own the least of the project but absorb most of the noise and much of the downside.
“The bad news is that the people who live next door own the least of it, they get most of the noise and very little of the upside, and if it all goes wrong, they will probably get a share of the losses too.”at 22:42
Big local incentive deals for factories and data centers often deliver far fewer jobs than promised.
Boyle gives two examples of large local incentive deals that fell short. In Mount Pleasant, Wisconsin, a village of 28,000 spent hundreds of millions on roads, water lines and tax incentives for Foxconn, which promised 13,000 jobs. The result was about a thousand jobs and the infrastructure bill. New York State spent almost a billion dollars on a Buffalo factory for SolarCity, and its controller later found about 54 cents of economic benefit per dollar spent. Boyle uses these cases to suggest that the data center subsidies now being offered carry similar risks.
“Foxconn then more or less changed its mind, and the village ended up with about a thousand jobs and the bill for the infrastructure.”at 23:26
AI buildout risks railroad-style overbuilding; current demand exists only because AI is nearly free.
Boyle weighs the case that AI is not a bubble. Labs are rationing compute, which suggests demand exceeds supply. He argues that this demand is at today's prices, and for most users the price is zero, while the companies spend far more than they earn. Nobody knows how much demand would remain if everyone paid full cost. He draws on a railroad analogy: each company can be sensible alone while the combined result is wild overbuilding, and six hyperscalers plus the big chipmakers make up about 20% of world stock market value. He cites an estimate that the buildout will cost 3.6% of US GDP a year until 2032. Boyle's conclusion is that the risk of overbuilding is real.
“The catch is that this is demand at today's prices, and for most people today's price is zero.”at 27:00
AI infrastructure & data centers
Local data center moratoriums won't slow AI much; projects just relocate to permissive states.
Boyle cites Princeton professor Arvind Narayan, who calculated that an average one-year state-level moratorium would slow AI capability growth by only hours. He concludes the projects do not stop; they move to a state that has not banned them. Ireland shows the same pattern: the grid operator stopped connecting new Dublin data centers, but developers adapted by bringing their own power and buying renewables. The build-out continued. Boyle therefore expects local moratoriums to have little effect on the overall pace of AI infrastructure.
“This is also why the local moratoriums probably won't slow things down much.”at 27:29
Shows
- Patrick Boyle · 2 episodes
Ask your AI what Patrick Boyle thinks
Chorus gives Claude, ChatGPT and other assistants every view Patrick Boyle has shared on the shows it covers, next to everyone else's. Ask where they stand, what changed, and who disagrees.
Chorus reports what speakers said. It isn't investment advice.