> ## Content Index > Fetch the complete content index at: https://longtalkshort.com/llms.txt > Use this file to discover other available public pages before exploring further. # How the Bursting of the AI Bubble Will Save Bitcoin - URL: https://longtalkshort.com/how-the-bursting-of-the-ai-bubble-will-save-bitcoin/ - Published: 2026-10-09T00:00:00.000Z - Updated: 2026-10-10T03:15:49.000Z - Description: The AI capital race looks like Bitcoin's rival: its bonds pay interest, its stocks sell growth, and both are draining the money that used to chase risk. But the same debt mountain may be the strongest case yet for owning an asset that is nobody's promise. - Author: Truman - Tags: Finance, #Import 2026-10-10 03:13, Richard Detente *A Bitcoin analyst's counterintuitive thesis: the trillions of debt financing AI are a headwind for Bitcoin today, and the very reason to hold it tomorrow. Distilled, with the speculative parts named.* Three headlines, no obvious connection. This week the US 10-year yield pushed past 5%, a level unseen since 2007\. Last month the Treasury sharply increased its buybacks of long-dated debt to support a market under strain, tripling the size of some operations in early September with little success so far. Meanwhile Alphabet announced an $84 billion equity raise to fund its data centers, and Amazon had to dangle a premium to place $25 billion of debt. And Bitcoin, for all its August recovery, still sits below where it started the year. Grand Angle Bitcoin argues these three facts tell a single story. We all watched AI "threaten" Bitcoin, in very large quotation marks, through a supposed rivalry over mining. But the more serious competition may not be happening in the server farms. It is happening in the portfolios. The tech giants are offering investors two things, a loan that pays interest or a share of potentially enormous future profits, and against those two propositions, risk assets have to defend their place. The twist, and the reason for the title, is that this very same movement could eventually become one of the best reasons to own Bitcoin. To see why, you first have to understand why, before that day of reckoning, the opposite is likely to happen first. This is a video essay from a Bitcoin-focused channel, so read it as a committed bull's argument, carefully built but not neutral, and with its speculative steps, which the narrator openly admits, flagged along the way. ## Where the AI money actually comes from Building the infrastructure behind your favorite large language models costs a fortune, and it costs it now, while the profits are expected later. The scale is staggering: capital expenditure and R&D by the hyperscalers went from $460 billion in 2024 to $870 billion over the last twelve months. At that level, even Amazon, Microsoft and Alphabet find their cash flows no longer enough, so they increasingly turn to debt, with cash flow becoming the war chest that services the interest and lets them borrow still more. Here is the number that matters. If you look only at long-dated debt and convert everything into 10-year equivalents (adjusting different maturities for equal interest-rate risk so they can be compared), these hyperscaler issues represent roughly 45% of the new long-term borrowing the US Treasury itself is doing this year. Even with that technical caveat, the figure is enormous, because these companies are arriving with massive financing needs at exactly the moment the American state also has to find buyers for its own debt. They are, in the end, reaching into some of the same pockets. Put yourself in a bond manager's shoes. You can lend to an American state whose deficit drift everyone is commenting on, or you can lend to Microsoft or Amazon, who carry the future. For part of your portfolio, the second proposition is tempting. Except that, by arriving en masse, this debt loses its scarcity. Since January, the yield spread on hyperscaler bonds has widened by about 30 basis points, against just 2 for the rest of the investment-grade market. In July, Amazon had to offer nearly 20 extra basis points on its longest maturities to place its issue. ## When AI raises the price of capital This is where the competition for capital becomes a macro subject. More long-dated debt to absorb, more choice for buyers, and additional pressure on the return they demand. The hyperscalers are obviously not the only cause: the war with Iran sent oil back above $100, US inflation remains high, and the Fed was preparing to raise rates for the first time since 2023\. But the rise in corporate debt issuance is now, in black and white, on the list of causes the financial press cites for higher rates. To be complete, not everyone agrees on the causality. The effect of AI on rates does not run only through the debt needed to finance it. PIMCO observes that around the big model releases, anticipated long rates actually fell, because the market can also see AI as a source of future productivity but also of economic uncertainty, which can raise demand for safe assets. In the short and medium term, though, the observation stands: to finance the race, a mountain of long debt has to be absorbed, and if that pressure keeps long rates high, the consequences reach far beyond the bond market. To pull your money toward a risky asset, someone now has to offer you a prospect attractive enough to beat what you can already get with much less risk. Why take a lot of risk, in other words, if a prudent allocation already pays you close to 5%? ## Why high rates punish Bitcoin You might say you would rather hold an asset that appreciates sharply than a placement that pays you a few percent a year. Fair enough, and the narrator agrees. But you have to understand the constraints of a large part of institutional finance. An individual building wealth can mostly watch what the investment will be worth in five or ten years. A pension fund has committed to paying retirements on precise dates, to beneficiaries with expenses every month who will not wait for the market to recover before paying their bills. There is also a whole category of funds built to pay monthly or quarterly distributions, chosen by their investors precisely to receive money regularly. For these players, bond coupons and repayment dates let them build a calendar and avoid selling assets at the wrong moment to raise cash. A hoped-for price rise several years out does not serve the same function as a payment scheduled for a given date. So when bonds pay more, part of this investor base simply needs less exposure to risk to hit its targets. Bitcoin can rise, but while you wait for that rise, you give up the interest your capital could earn elsewhere, and the more attractive that alternative, the heavier the sacrifice. That is opportunity cost, and it is not theory: when rate-hike expectations climbed in early September, the financial press explained Bitcoin's drop precisely by the fact that it pays no interest. An IMF working paper finds the same mechanics, with Fed tightening weighing on Bitcoin through the risk-taking channel, exactly as it does on equities. None of this is mechanical. In August, Bitcoin gained nearly 25%, carried by more than $3.5 billion of inflows into the spot ETFs, even as the hyperscalers kept issuing. Its price depends on other forces that can more than offset this effect. But all else equal, the costlier capital is, the less favorable the environment for risk assets. It is a headwind, not a sentence to fall. ## Can the tech giants vacuum up all the capital? The hyperscalers' sorcery does not stop at making bonds look attractive. The same giants could also capture a good share of the money that still wants risk. Suppose their pharaonic investments prove genuinely, durably profitable, earning more than the capital they consume costs. From there, a virtuous circle unique to these players can take hold. They issue bonds; the money funds data centers and compute; those investments generate activity, revenue and extra profit; that value flows to shareholders; earnings prospects rise and support the share price; and the success reassures lenders, who see borrowed capital turned into profit and therefore into repayment capacity. JP Morgan describes exactly this link between return on investment, shareholder value and credit quality. That confidence supports demand at the next issue, fresh capital funds new investment, and the loop continues. This is how these companies can win on both tables, attracting investors who want a contractual return while capturing those who want to ride the growth. Keep in mind, though, that the loop is not guaranteed. It holds only as long as the return on invested capital stays above its cost. Still, as long as Bitcoin is bought mainly as a bet on its own price going up, it lands squarely in this competition: on one side bonds that pay interest, on the other stocks carried by the prospect of considerable profit, and in the middle a volatile would-be haven. Compared with unappealing government debt, Bitcoin looks reasonable; compared with the healthy debt of genuinely expanding companies, perhaps less so. Presented this way, Bitcoin seems to lose twice. Except it holds one advantage neither of the others can offer, and this is where the whole story changes meaning. ![](https://storage.ghost.io/c/d7/93/d793216c-cf13-441f-bdae-4f3aef998c0e/content/images/2026/10/four-promises-1.png) ## The advantage Bitcoin alone holds Bitcoin sits in the risk-asset box today because it is still in its adoption phase. But its original proposition was never to be a tech stock. The long-term thesis, the one now playing out, is that of a reserve asset, which, incidentally, is what Hal Finney foresaw. If that bet holds, Bitcoin simply stops playing in the same category, because at that point the rise in rates driven by ever-stronger demand for capital would no longer be a headwind for it. It would become the symptom of a system in which companies and states must issue ever more debt to fund their needs, a world of capital scarcity. And the nature of what is being financed suggests the timing lines up better than it first appears. Those issues are massively long-dated for a reason: data centers, substations and grid connections amortize over decades. The commitments created today will still be here when Bitcoin has perhaps finished its adoption phase. Yes, the narrator says, this is speculative, and he owns it. But the more this debt swells, the more the question becomes where to store capital without depending on any issuer's balance sheet. A Microsoft bond is a promise from Microsoft. A Treasury is a promise from the American state. An Alphabet share is a claim on Alphabet's future profits. Bitcoin is the only proposition on the list that rests on no particular party's promise to repay. That is why he calls it everyone's debt: a debt not forcibly collectible, certainly, but a debt all the same. Gold is the precedent. Central banks adore it. In the World Gold Council's 2026 survey, a record 45% of them plan to increase their own gold reserves over the next twelve months, and gold, he argues, is just Bitcoin but worse. What interests him most is what came before. For more than twenty years central banks were net sellers of gold; in the 2000s the yellow metal was seen as an anachronism precisely because it earned nothing in a financial universe built on interest-bearing debt. It took the 2008 crisis for them to turn net buyers again in 2010, for the first time since 1989\. Gold's safe-haven qualities express themselves when monetary distrust returns, and what changed was confidence in issuers' promises. BlackRock already wrote in 2024 that a substantial part of institutional interest in Bitcoin came from worries about US debt and deficits, and in late August 2026, with federal debt having just passed $40 trillion, BlackRock's head of digital assets repeated that these debt levels push part of the investor base toward Bitcoin and gold. ## Why Bitcoin is still, for now, a risk asset So how do you break the risk-asset illusion? When the war with Iran erupted, Bitcoin did not behave like a haven. It broadly followed risk appetite and fell, because its correlation with equities tends to climb precisely in phases of stress. Gold fell over the same period too, true, but gold had already gained 60% in the eight months before the strikes; the geopolitical risk was already in the price, and traders sold the event after months of buying the rumor. Bitcoin had no such premium to give back, since it entered the war already well below its record, and its moves simply hugged risk appetite. Even the governor of the Czech central bank, one of the central bankers most open to Bitcoin, describes it as an asset with characteristics closer to venture capital. And the rise in Bitcoin's correlation with equities coincided, tellingly, with the arrival of institutional investors, though which institutions matters, as we are about to see. Leaving the risk-asset box will not be decreed. It will be built gradually, and the levers lie, the narrator argues, in the balance sheets of those who hold Bitcoin and in the rules that govern those balance sheets. ## How Bitcoin can become a true reserve asset First, in a liquidity crisis you do not sell what you want, you sell what you can, and Bitcoin, open 24/7, is often the first window still trading. As long as a large share of the supply is in the hands of leveraged players or arbitrage strategies, it behaves like a risk asset. We saw it in June: $2.5 billion left the spot ETFs, but most of it came from arbitrage strategies while passive holders kept buying. That is the whole difference between an investor and a saver. The saver with no debt opposite the position does not sell to cover a margin call. The more the supply shifts to that kind of holder, the less forced selling weighs in a crisis. Then comes volatility, which compresses as the market deepens. For a reserve manager, volatility is pure risk, the constraint that decides whether an asset may enter the portfolio at all, and the trend is real: across 2025, Bitcoin was less volatile than Nvidia. But a falling average volatility does not erase 50% drawdowns, and as long as those exist, Bitcoin stays out of reach of most reserve mandates. There is also the regulatory plumbing, because a box is also a legal label. Since 1 January 2026, the Basel standard applies a 1,250% risk weight to Bitcoin on bank capital, and as long as that rule holds, no bank can treat Bitcoin as a reserve asset. But the Basel Committee has launched an accelerated review of the standard, with an update announced for this year, and the United States has signaled it does not intend to apply it as written. Finally, the quality of the institutional buyer matters. In November 2025 the Czech central bank bought about $1 million of Bitcoin, stablecoins and tokenized deposits in a test portfolio, and its governor presented an analysis showing that 1% Bitcoin in its $180 billion of reserves would raise expected return without materially changing overall risk. Tiny, yes, but that is exactly how gold regained its place, not all at once but central bank after central bank. And this kind of actor has nothing in common with the arbitrage funds from before. A central bank has no margin calls, no clients demanding their money, no spread to capture. It holds a reserve asset for its function, over a horizon of decades, so it has no mechanical reason to sell when the price falls, as gold showed again this year when central banks held to their buying plans despite a drop in the price. ## The test that could flip everything Then there is the real test, the one that will tip the box or not. Distinguish two kinds of crisis. In a liquidity crisis, everything liquid is sold, Bitcoin included. But in a crisis of confidence in an issuer, the story can differ. In March 2023, when Silicon Valley Bank collapsed, Bitcoin gained nearly 40% in about ten days, even if expectations of Fed support also played a part. The day Bitcoin holds through a genuine crisis of confidence in a major issuer's debt, and not just through a bank panic, the market will probably start filing it somewhere else. With a 10-year yield above 5% and a Treasury buying back its own long-term debt to contain its rates, that test may not be as theoretical, or as far off, as it sounds. The point is not that the AI race leaves Bitcoin's rivals looking more attractive, though it does: their bonds pay and their shares offer growth. It is that this same race has built a world where financing needs and debt commitments take ever more room, where even the most solid companies on the planet must pay more to borrow, and where the most powerful state in the world must buy back its own long-term debt to try to hold its rates down. In that world, holding an asset with no issuer can acquire a very particular value. The whole demonstration rests on one question. Will Bitcoin stay a risk asset that pays nothing, or will it end up becoming the reserve asset it claims to be? Bitcoin does not need to win the yield race. It needs to still be there, intact, the day everyone asks who is really able to repay. ## How to read this Treat this as a thesis, not a forecast, and weigh it against its source and its own admissions. It comes from a Bitcoin channel and reads like one: the framing that "gold is just Bitcoin but worse," the confidence that the reserve-asset thesis "is now playing out," and the glide from headwind to tailwind are advocacy, and the narrator concedes the core of it is speculative. Several of the figures are forward-dated or come from the video's own reading of events (a $40 trillion federal debt, a 2026 World Gold Council survey, the Basel 1,250% weight, the Czech central bank's test), so check them before repeating them as settled fact, and note that the sources the piece leans on hardest (BlackRock, JP Morgan) are themselves building Bitcoin and AI businesses. What is genuinely clarifying, whatever you conclude about the price, is the distinction at its heart. There is a difference between an asset that is a claim on some issuer, a bond, a Treasury, a share, and an asset that is a claim on no one. Most of the time, in a calm world of interest-bearing debt, being nobody's promise is a weakness, which is why Bitcoin pays you nothing and gold sat ignored for two decades. The argument here is that a world drowning in long-dated debt is exactly the world in which that weakness turns into the point. You do not have to buy the conclusion to find the question worth holding onto: in a crisis of confidence, who can actually repay, and what do you own that does not depend on the answer? --- ## Go deeper - [The Bitcoin Standard: The Decentralized Alternative to Central Banking](https://www.amazon.com/s?k=The+Bitcoin+Standard&ref=longtalkshort.com), Saifedean Ammous Affiliate links, a small commission supports the blog at no cost to you.