Archived reading, published Thu, 27 Aug 2026 14:19:31 UTC (25 days ago). This is not the current state of the meter.
See the live reading →Crash Lab watches the machinery under the current boom: the debt paying for AI data centres, private credit, the leverage that has moved out of banks into places nobody has to mark, the bond market and the dollar.
Every four hours it reads the day’s reporting from 19 sources and rewrites this page.
Status: Held at 61. Three genuinely new things cancelled out: Nvidia added $160bn of supplier purchase commitments in a single quarter (new tinder), but its demand assumption was tested this quarter and held, and the bitcoin-treasury model is visibly deleveraging, with the top 50 companies' combined value down from $150bn to $67bn and many selling coins. The tape is the same calm we already counted — VIX 14.81, high yield 270bp, investment grade 81bp, S&P 1.2% off its high.
Reporting from 26 Aug to 27 Aug
Eight places it could go, scored 0–10. Tap one for the explainer.
What changed in the last few hours, and what each one says about the plumbing.
The vendor at the centre of the AI build has now written large obligations to its suppliers and its customers simultaneously, so a demand shortfall hits it from two directions at once.
The reporting arrived five weeks after the loss, which is how long it takes to find out what a large non-bank was actually holding.
A 55bp premium on the strongest borrowers in the market is the cleanest available price for AI risk, and it is being paid quietly while headline credit spreads stay tight.
A leveraged structure genuinely unwinding is one of the few things that lowers fragility, and this one is unwinding into a rising coin price, which is the gentlest way it could happen.
The metric that captures borrowers already in distress is deteriorating far faster than the one managers report, and BDC share prices are near highs.
Using the government's cash buffer to hold down long yields converts a debt-management tool into monetary policy, run by the branch that is not supposed to set it.
Stuff you wouldn’t have believed was possible until 2026.
Leopold Aschenbrenner, 24, married in August at the end of a week in which heavy losses forced the sale of his hedge fund's positions. Among the guests was Rob Granieri, co-founder of Jane Street, which had invested in the fund and had just posted its first losing month in a decade — a $15bn hit to trading revenues.
Of the 50 largest corporate bitcoin holders — including a Spanish coffee shop chain, a Japanese clothing retailer and a US battery maker — 43 now trade below their price before they announced the crypto pivot, and 35 have lost at least half their value.
QTS Realty sold $3.9bn of bonds to fund a Georgia data centre tied to Microsoft at a yield of about 7.23% — despite an investment-grade rating. BlackRock paid 7.53% in July on blue-chip securities for a Texas project. The whole US investment-grade index trades at 81bp over Treasuries.
Ehime Shinkin Bank on the island of Shikoku lost ¥18.9bn in the year to March, its first annual loss since it was founded in 1951, on Japanese government bonds bought during the ultralow-rate years. Seventeen shinkin banks were loss-making, up from five, and 237 of 245 are carrying unrealised losses on securities.
What the crash-callers are saying, checked against real reporting.
“Nvidia's CFO commentary discloses $105bn of guarantees on the OpenAI Ohio facility plus an option on 3.8 more gigawatts, which could bring total off-balance-sheet guarantees to about $198bn.”
The disclosed parts check out: guarantees capped at $105bn, twenty-year leases to OpenAI, phasing in from fiscal 2029, and an option to provide credit support for roughly 3.8 additional gigawatts. The $198bn total is an extrapolation from the optional tranche, not a disclosed figure, and Nvidia has not put a number on the option.
Claimed by Meet Kevin
“Nvidia wants to give OpenAI a $125bn guarantee.”
The announced figure is $105bn. Earlier talks reportedly ran to $250bn before being cut back; the WSJ put Nvidia's revised exposure at under $120bn. Nobody has reported $125bn.
Claimed by Wealthion
“The Federal Reserve has bought over $330bn of Treasuries in the last twelve months — QE by another name.”
The Fed's own H.4.1 shows Treasury holdings up roughly $325–340bn year on year, so the number is right. Whether it is QE is a labelling argument: the Fed describes it as an end to runoff and reserve management, and it is not an announced purchase programme with a target.
Claimed by Palisades Gold Radio
“Central banks are adding around 1,000 tonnes of gold a year for the fifth consecutive year.”
True for 2022, 2023 and 2024. 2025 came in at 863 tonnes, and World Gold Council data put first-half 2026 at 345 tonnes with 289 tonnes in the second quarter. Official-sector buying is large and persistent, but the streak above 1,000 tonnes has already broken.
Claimed by Palisades Gold Radio
Dispatches from previous readings. The same argument, no longer the news.
The vendor is no longer just guaranteeing its customers' rent. It is signing leases itself on the assumption it can sell them to a buyer base it is currently helping to invent.
A single buyer that does not care about price is now the swing bidder in the world's second-largest government bond market, and its buying is conditional on stocks going up.
Default rates across the four largest private lenders are at five-year highs while their listed vehicles trade at their highs. The gap is entirely a question of when the prices they put on their own loans catch up.
The regulatory repair for the 2020 government bond blow-up does not cover the two structures used to finance the largest leveraged positions in the market.
The channel by which digital tokens pegged to the dollar could drain bank deposits is being built out by governments at the moment the underlying usage has stopped growing.
The market is already discriminating between AI spending funded by cash flow and AI spending funded by borrowing, before any broad measure of corporate borrowing costs shows stress.
Gates are how a problem with loans turns into a problem with cash, without a single borrower failing to pay.
The new cloud-computing funding model turns customer concentration into funding concentration, and neither shows up as borrowed money on anyone's books.
The model where a company buys digital assets using index-fund demand depends on index membership it does not control, and the index provider is now asking whether it should be there at all.
Investors are buying insurance against the policy mix without selling the currency, which tells you they expect erosion, not rupture.
If the Treasury Department is setting the long-term interest rate, the Federal Reserve's inflation target becomes a statement of intent rather than a constraint.
Household stress is not spreading. It is concentrating, and concentrated losses land on specific lenders rather than the economy.
This is the first time a national securities regulator has said publicly that private credit funds in its jurisdiction are blocking withdrawals while borrowers fail. That is the sequence everyone has modeled and nobody has watched happen.
If the chips are not real security for the loans and insurance is not available, the entire $7 trillion is unsecured lending to four or five tenants, dressed up as secured.
Japan's government bond market at the long end is being held up by one rebalancing rule at one pension fund. That rule flips direction if stocks fall.
The trade where hedge funds borrow heavily to profit from a tiny gap between two nearly identical prices in the US government bond market is now bigger and more concentrated than it was in 2020, and the reform designed to contain it has stopped expanding.
Household credit is splitting into two populations rather than deteriorating as one, which means the overall numbers will keep looking calm right up until the bottom third stops paying.
If the Treasury Department starts using its cash balance to push down long-term interest rates, debt management and monetary policy have merged, and only one of them has an inflation mandate.
The specific, observable things that would move the number - in either direction.
The expanded buyback programme starting on 9 September and failing to hold the 30-year below 5.2% — an intervention that is tested and does not work would move ignition, not fragility.
Would move the number
A named US semi-liquid private credit fund formally gating redemptions, or a BDC marking its book down more than 5% in a quarter; so far the gating we can source is one Australian fund and an aggregate dataset.
Would move the number
High yield through 350bp or investment grade through 100bp — at 270bp and 81bp, credit is telling us none of this is a problem yet, and we would want to know the moment it changes its mind.
Would move the number
A hyperscaler cutting capex guidance, or Nvidia's supplier commitments shrinking next quarter — that would be a genuine fall in fragility, and we would mark it down.
Would move the number