AI Bubble Tracker

“The dependency of the hyperscalers on Anthropic and OpenAI is just huge — both lose billions and rely on raising capital for survival. Oracle's debt is rated barely above junk. There is no hedge for the LLMs: hyperscalers have fallback franchises, the labs don't.”

“First full read across all four hyperscalers: Microsoft's commercial backlog jumped to $678B (from $625B), closing in on Oracle's $638B. Google Cloud posted $155B (+82% YoY) and Amazon's AWS posted ~$200B — both reported here for the first time. Backlog growth is not just an Oracle story anymore; it's broad-based across the sector.”

“No fresh backlog figures this week — Microsoft, Google and Amazon already reported their quarters, and Oracle's next earnings (fiscal Q1 FY27) isn't until Sept 9-12, which is this signal's next real checkpoint.”

“Still no new backlog print — Oracle has now confirmed its fiscal Q1 FY27 earnings date as Thursday, Sept 10 (after market close), so the number itself remains four days away as of this sweep. The wait is happening against a noisier credit backdrop: Oracle shares have continued sliding (down roughly 30% over the trailing month per multiple trackers) on investor worry about its debt load and its concentration in a single, unprofitable customer (OpenAI is reported to account for over half of the $638B backlog). None of that changes the backlog NUMBER, but it raises the stakes on whether Sept 10's print shows growth continuing or cracking.”

“Oracle reported Sept 10: backlog hit a record $664B, up $209B year-over-year — but the QUARTER-OVER-QUARTER pace slowed sharply, from about $85B of new backlog added last quarter to about $26B this quarter (roughly 46% YoY growth this print, down from the ~363% YoY comparison logged a quarter ago). That is the first real sign of deceleration this signal has shown since this tracker started — moving status from 'accelerating' to 'watch' until next quarter's report (expected around December) either confirms a genuine slowdown or shows it bouncing back. Oracle's own capex jumped to $28.5B for the quarter (from $8.5B a year ago) and it raised $20B in a stock sale, both signs it's still spending hard against that backlog. The stock popped as much as 8.5% right after the report, then reversed to close down about 1.7% — investors cheered the headline number, then had second thoughts. No update from Microsoft, Google, or Amazon this week (their next earnings are late October).”

“Quiet week — no fresh backlog figure from Oracle, Microsoft, Google, or Amazon (all four have already reported their most recent quarters; the next round is late October/December). Nothing found this week to confirm or reverse the deceleration flagged last week. Holding at 'watch' pending Oracle's next report.”

“Still no fresh backlog print. Oracle's next scheduled event is an Oct 28 Financial Analyst Day (not a full earnings release with a new RPO number) — its next quarterly report isn't until Dec 14. No new figures found from Microsoft, Google, or Amazon either. Holding at 'watch' pending a real new print; nothing this week confirms or reverses the deceleration flagged Sept 13.”

“Still no new backlog figure (Oracle's next RPO print is Dec 14; its Oct 28 analyst day is not an earnings report). Two quality signals emerged on the backlog's biggest counterparties, though. (1) Bloomberg (Sept 24-25): Oracle sent a force majeure notice to Blue Owl, the developer of Project Jupiter, the 2.45 GW Stargate campus in New Mexico. Pipeline permits were denied twice and the gas line was pushed back to Feb 1 2027. The notice lets Oracle delay payments if the site misses its 2028 target. Oracle says it remains committed and on schedule. This is the first time a flagship OpenAI-linked site has formally invoked delay protections. (2) Anthropic's prospectus (per Reuters, Sept 28) lists $518B of future cloud, compute and infrastructure obligations, which is largely the hyperscaler backlog seen from the other side. It also warns that its largest customers aren't locked into long-term contracts. Neither fact changes the backlog totals, but both show that the 'large' backlog depends on power permits and on unprofitable borrowers. Holding at 'watch'.”

“No new backlog figure this week. Oracle's next report isn't expected until around Dec 14 (its Oct 28 analyst day is not an earnings release); Microsoft, Alphabet and Amazon's Q3 reports are expected in the next one to two weeks (Amazon is reported, unconfirmed, for Oct 29) and are this signal's next real checkpoint. Holding at 'watch'.”

“Still no new backlog figure — Microsoft, Alphabet, Amazon and Meta all report Q3 earnings Oct 28-29, now just days away, which is this signal's real next checkpoint; Oracle's next print remains Dec 14. Holding at 'watch'.”

“Nvidia's $500B securitized-financing push means some data-center capex shifts off hyperscaler cash flow — hyperscaler FCF 'could improve' optically while the debt stack extends into banks and private credit. Read improving FCF with suspicion from here.”

“Kill-trigger watch candidate: Meta is getting squeezed — latest quarter revenue +28% but expenses +55%, no cloud business to monetize AI through. Most likely first major player to cut; watch how the market reacts when it does.”

“The five biggest spenders (Amazon, Alphabet, Microsoft, Meta, Oracle) now guide to a combined $775-800B for 2026 — roughly 3x the ~$238B spent in 2024. This is the opposite of the kill trigger so far: Alphabet RAISED its 2026 capex guidance (to $195-205B) and its stock fell about 7% over the following week — a capex increase punished, not a capex cut rewarded. That's still consistent with the current framing (investors want the spending to keep climbing), just a reminder the market is getting twitchier about how big the numbers are getting.”

“Still no kill trigger, and the numbers keep climbing: Nvidia's Aug 26 earnings (record $96.2B quarterly revenue, Data Center alone $89B, Q3 guided to $108B) had its CFO reiterate that the top hyperscalers' combined capex is heading toward roughly $1.3T next year, up from about $800B now. But the same 'raise-without-proof gets punished' pattern repeated: Meta fell about 10% on a JPMorgan downgrade after guiding 2026 capex up to $125-145B, the second capex raise in three weeks (after Alphabet) to get sold off rather than bought. Microsoft and Amazon, whose capex raises are tied to demonstrated cloud revenue growth, are not seeing the same reaction — the market is starting to price the difference between spending that's paying for itself and spending that isn't.”

“No new capex guidance this week — none of the five biggest spenders reported earnings (the next round lands in October for Q3). Still no kill trigger. The clearest sign spending keeps translating into real infrastructure: OpenAI said its new GPT-6 Astra model, released Sept 3-4, was trained on more than 100,000 GPUs at its Stargate site in Texas — by far its largest training run — which is capex showing up as product rather than as a guidance number. Separately, investor Michael Burry kept up his public campaign arguing the whole buildout rests on circular financing, this time estimating roughly $879B in hyperscaler commitments circling through Nvidia; that's a credibility challenge to the spending story, not evidence of an actual cutback.”

“Oracle's Sept 10 earnings were this week's only real capex data point: it spent $28.5B in the quarter (versus $8.5B a year ago) and held its full-year guidance flat at $90-95B gross capex — not a cut. Still no kill trigger anywhere: nobody announced a spending cut, let alone had their stock rewarded for one. Oracle's stock actually did the opposite of the kill-trigger pattern — it rallied on the spending/backlog news, then gave the gain back within the same session, which reads more like investors getting nervous about HOW the spending is funded than about the spending itself.”

“No hyperscaler updated capex guidance this week (next earnings late October). Checked specifically for the kill trigger (a capex cut rewarded with a stock rally) and found none — commentary this week was speculation that 'the next leg will be a big hyperscaler announcing a capex cut,' which is market anticipation, not an actual event. Kill trigger still not fired.”

“A capex tracker roundup (covering the trailing four reported quarters through mid-September) confirms the pattern holds: every major hyperscaler in the tracker raised or held its guidance this cycle — none cut. The four biggest (Amazon, Microsoft, Alphabet, Meta) spent a combined $543B of the group's $657B total over that period, funded from operating cash flow. No capex cut of any kind was announced this week, so the kill trigger (a cut that gets REWARDED with a stock rally) still has not fired and there is still no candidate event to even assess. Status holds at 'accelerating.'”

“No hyperscaler capex guidance changes this week. Big Tech's Q3 earnings are the next real checkpoint — Amazon is reported (unconfirmed) for Oct 29, with Microsoft, Alphabet and Meta expected in the same window; Oracle's next report isn't until Dec 14. No kill-trigger candidate event found. Status holds 'accelerating'.”

“No capex cut announced anywhere, so the kill trigger (a cut REWARDED with a stock rally) still hasn't fired. The narrative did shift slightly: UBS now projects hyperscaler capex growth slowing from +76% this year to +25% in 2027 and +6% in 2028, and some fund managers are reportedly rotating out of chip stocks into hyperscaler equity — but both are forecasts/rotation, not an announced cut. Big Tech's Oct 28-29 Q3 earnings, with first real 2027 guidance, are the actual test. Status holds 'accelerating'.”

“Red here does not mean Chinese models are failing — it means pressure against US share is RISING. This signal measures the force pushing on US revenue speed.”

“Independent corroboration: enterprises are using Chinese open-weight models more and more; 'overcharging for tokens is going to be very difficult with good-enough competition.' A price war is possible — and Anthropic + OpenAI are ~70% of hyperscaler AI revenue, so a price war transmits straight into the backlog.”

“New Chinese releases kept coming this week (Alibaba's Qwen3.8 Max Aug 2, Z.AI's GLM-5.2 Turbo Aug 17), but the readings on actual Chinese-model traffic share are noisy right now — trackers show anywhere from ~45% to over 60% of OpenRouter volume depending on whether you measure the top 10 models or all traffic. Treat the exact percentage as uncertain this week; the direction (more competitive Chinese releases, more enterprises adopting them) is not.”

“More Chinese releases (Z.AI's GLM-5.3-Flash Aug 26) and readings still cluster high, 51-61% of OpenRouter traffic by some trackers, with DeepSeek the single largest provider at roughly 16%. One genuine crack in the 'nearly free' narrative: DeepSeek's new V4 Pro model raised prices roughly 14x with new peak/off-peak billing — the first sign a leading Chinese lab thinks it can charge more, not less. Moonshot's Kimi K3 reportedly had to pause new subscriptions due to demand. Keep treating the precise percentage as noisy; the competitive pressure itself is not in question.”

“Alibaba's Sept 2 refresh of Qwen3.8-Max took the #1 spot on a major coding leaderboard while leaving pricing untouched — the opposite of DeepSeek's recent price hike, and a reminder that the 'nearly as good, nearly free' pressure this signal tracks hasn't gone away just because one Chinese lab tested higher prices. On the US side, OpenAI answered competitively with GPT-6 Astra (Sept 3-4), its most capable model yet — the two labs are clearly racing each other as much as they're racing the cost pressure from China. Exact OpenRouter share numbers remain too noisy across trackers to log a new reading this week; the direction of competitive pressure is unchanged.”

“Quiet week for new Chinese model releases — nothing dated this week from Moonshot, DeepSeek, Alibaba, or Z.AI. One useful data point resurfaced: a Sept 1 snapshot puts Chinese-origin models at roughly 46% of identified enterprise token usage (a different measure than the OpenRouter US-vs-China split tracked above, so not logged as a new point on that same series, but directionally consistent with 'pressure keeps building'). On pricing, OpenAI's new GPT-6 Astra actually launched ABOVE its prior model's price ($10/$50 per million tokens) rather than cutting — worth noting as a case where the US labs are charging a premium for their best model instead of racing China to the bottom, even as the volume-share pressure keeps building underneath.”

“No confirmed brand-new frontier Chinese model release strictly this week (closest are GLM-5.3 and DeepSeek V4, both dated right at/before the window edge). One secondary aggregator (LinkedIn-sourced, not OpenRouter itself) claims Chinese models hit 61% of OpenRouter token volume for the week of Sept 7-13 — flagged as unverified, not logged as a confirmed new reading on the usSharePct series held at 30%. A genuine countertrend worth noting: DeepSeek has reportedly raised its own API prices on some tiers by as much as 12x, the opposite of the 'nearly free' pattern this signal watches for — a second Chinese lab (after DeepSeek V4 Pro's August hike) now testing whether it can charge more rather than less. Meanwhile Anthropic launched Opus 5 at roughly half its prior flagship's price while holding Sonnet 5 flat, so competitive pricing pressure is coming from both directions this week, not just from China.”

“No major new frontier Chinese model release found strictly this week. Several secondary aggregators (not OpenRouter's own dashboard) put combined Chinese-origin share of OpenRouter token volume around 44-46% currently — directionally consistent with the pressure this signal tracks, but methodology differs enough between trackers that it isn't logged as a confirmed new reading against the 30%/70% US/China split already on file. The bigger news this week is actually a US-vs-US price war, which matters for this signal because it shows the pressure Chinese models created is now also being self-inflicted: OpenAI launched GPT-6 Sol and Luna (Sept 22-23) — Sol priced at half of Anthropic's new Opus 5.5, and Luna at roughly 1% of OpenAI's own Astra price — right after Anthropic shipped Opus 5.5. Read together: the 'nearly as good, nearly free' dynamic this signal was built to watch for China is now also happening between the two big US labs, which arguably makes the overall revenue-per-token squeeze even faster-moving than a single new Chinese release would. Status holds at 'decelerating' (pressure still rising).”

“No confirmed new frontier Chinese model release found this week. Share estimates remain as noisy as ever across trackers — roughly mid-40s% to 60%+ of OpenRouter token volume depending on methodology, with DeepSeek still the single largest individual provider (~16%). Not logged as a new reading against the 30%/70% US/China split already on file; the direction of pressure is unchanged.”

“No confirmed new frontier Chinese model release found this week either. Trackers still disagree widely on the exact share (estimates range from roughly 35% to over 60% of OpenRouter/enterprise token volume depending on methodology and whether enterprise or total traffic is measured). Not logged as a new reading against the 30%/70% split on file; pressure direction unchanged.”

“Both labs are still accelerating, not just growing. Anthropic told investors its annualized revenue run-rate hit $65B in July — up 7x from a year earlier — and investors expect it to keep growing at roughly that pace, finishing 2026 around $100-120B. OpenAI's run-rate reached about $40B, roughly double where it stood in the spring. This is the strongest reading this signal has had — moved from 'watch' to 'accelerating' this week — but both companies are still deeply unprofitable, so the speed is what's propping up their valuations, not earnings.”

“No fresh run-rate print this week — largely a restatement with texture: Anthropic's preliminary Q2 revenue came in above $11.5B versus $787M a year earlier, and OpenAI's enterprise revenue passed its consumer revenue for the first time in July, with its paying customer count up 32% in a single month. Nothing here suggests deceleration; there's just no brand-new number to move the needle beyond what was already captured last week.”

“No new run-rate figure this week, but a genuine new revenue-generating surface: OpenAI rolled out GPT-6 Astra (Sept 3-4) — its most capable model yet, described internally as its largest training run ever — to ChatGPT Plus/Pro/Business/Enterprise, the API, and AWS, with API pricing set at $10/$50 per million input/output tokens. It's too soon for this to show up in a run-rate number, but a materially better product landing broadly is exactly the kind of thing that would show up in next month's read. Nothing suggests deceleration.”

“Quiet week — no fresh run-rate or revenue disclosure found from either company. Carrying forward Anthropic's ~$65B annualized run-rate (July) and OpenAI's ~$40B (August). Nothing found this week points toward deceleration.”

“Genuinely new number: NYT reporting (via Yahoo Finance, Sept 18) says Anthropic's full-year 2026 revenue is on track to top $100 billion — up sharply from the ~$65B annualized run-rate logged in July. This is the strongest reading yet for this signal; still accelerating. No new OpenAI figure found this week; its ~$40B run-rate (August) is carried forward.”

“Not a new revenue number, but a genuinely new and stronger quality signal: multiple outlets (Bloomberg/FT, SBS, Dealroom, dated from Sept 13 but confirmed and widely recirculated this week) report Anthropic told a small group of shareholders it expects to post an adjusted operating profit for a SECOND straight quarter, with gross margins above 80% before revenue-sharing payments to partners like Amazon. SemiAnalysis separately estimates roughly $1B of Q3 GAAP operating profit at a 6% margin. None of this is in an audited public filing yet — it's pre-IPO framing to shareholders, not a confirmed result — but if real it means the run-rate growth logged in prior weeks is increasingly translating into an actual (if adjusted) bottom line, not just top-line speed. No new OpenAI figure found this week. Still accelerating.”

“The strongest reading this signal has had yet: Reuters (Sept 29, same-day Bloomberg confirmation) reports OpenAI's annualized revenue run-rate is nearing $70 billion — up more than 70% since the start of Q3 alone, driven by enterprise sales roughly doubling since July and consumer revenue that outgrew the whole of 2025 within the quarter. That's a jump from the ~$40B run-rate logged in August, i.e. the growth RATE is itself speeding up, not just the level. No new Anthropic figure found this week; its ~$100B full-year 2026 projection (Sept 18) and expected second straight quarter of adjusted operating profit (Sept 13) stand unchanged. Still accelerating, and more clearly so than last week.”

“Correction to last week's reading: the $70B OpenAI run-rate figure (Reuters/Bloomberg, Sept 29) does not hold up. The Financial Times (Oct 8, with same-day Reuters/Zawya confirmation) reports OpenAI told investors its own annualized revenue was nearing $50B at the end of September — about $20B below the $70B figure that had circulated, which reportedly came from attempts by OpenAI's own investors to produce a direct comparison with Anthropic's numbers, not from OpenAI itself. $50B is still up from the ~$40B run-rate logged in August, so growth continues, but the stronger 70%-since-Q3 acceleration claimed last week is not confirmed. No new Anthropic figure this week; its ~$100B full-year 2026 projection and expected second straight quarter of adjusted operating profit stand unchanged. Status moves from 'accelerating' to 'watch' pending a cleaner read on OpenAI's actual pace.”

“No new actual run-rate figure, but OpenAI reasserted the bullish number as a forward target rather than a current state: Bloomberg (Oct 9) reports the company expects to reach or exceed $70 billion in annualized revenue by the end of 2026, driven by its expanding enterprise business — distinct from, and not a contradiction of, the FT's Oct 8 report that the actual September figure was nearer $50B. No new Anthropic figure this week; its ~$100B full-year 2026 projection and expected second straight quarter of adjusted operating profit stand unchanged. Status holds at 'watch' pending confirmation either number is actually reached.”

“No new run-rate or revenue disclosure found from either company this week. OpenAI's contested $50B (FT, actual) vs $70B (Bloomberg, year-end target) figures and Anthropic's ~$100B full-year 2026 projection stand unchanged. Status holds at 'watch'.”

“The deal now has names attached: Nvidia formalized memorandums of understanding with six of the biggest names in finance — Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs, and KKR — to raise the $500B+ in compute financing. Two new details sharpen the risk picture: the SEC issued guidance in July that removes a risk-retention rule which would otherwise have required sponsors to keep a slice of this debt on their own books, and Nvidia has personally guaranteed (with its own money) that the chips used as collateral will hold their value. Both moves make it easier for capital to flow into these deals and easier for the risk to end up somewhere other than where it was created — exactly the 2006 dynamic this signal is watching for.”

“Still MOUs, not signed final agreements, and still no remittance or delinquency data publishing. But the SEC guidance behind the deal (dated Aug 17 in wider coverage) got more scrutiny this week: it explicitly takes sponsors 'off the hook' on certain look-through requirements for data-center securitizations. BlackRock's Larry Fink compared the moment to the birth of the mortgage-backed-securities market in the 1970s — which, read straight, is either a compliment about unlocking capital or an omen, depending on how this decade's version ends.”

“Still MOUs only — no report found of the Nvidia/Apollo/BlackRock/Blackstone/Brookfield/Goldman/KKR arrangements converting to signed final agreements, and still no remittance or delinquency data publishing. The main new development is louder skepticism, not new structure: investor Michael Burry escalated his public case (including a Sept 4 post) that hyperscaler AI spending runs through a circular web of vendor financing, this time estimating that roughly $879B in hyperscaler commitments circle through Nvidia in ways that can inflate reported revenue without necessarily reflecting new end-customer demand. Nvidia's credit-default-swap spread has stayed elevated on this kind of scrutiny. None of this is new hard data on the securitization deal itself, but it's the same underlying concern this signal is built to watch — who ends up holding the risk when the labs paying the bills aren't profitable.”

“Still MOUs only — no signed final agreements and no remittance/delinquency data found. Burry himself made a move worth noting: on Sept 10 he disclosed he had trimmed his overall bearish positioning and let his December-2026 Nvidia and Palantir put options expire rather than rolling them into new contracts, while keeping longer-dated (2027) puts on Palantir and the Nasdaq (QQQ). He kept his three biggest short positions in place — Oracle, Palantir, and Nebius Group — so this reads as reducing near-term bet size heading into a heavy earnings/auction week, not as him changing his mind about the thesis.”

“No change this week — still MOUs only (Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs, KKR alongside Nvidia), no signed final agreements found, and no remittance or delinquency data publishing.”

“No new developments found this week — still MOUs only, no signed final agreements, no remittance or delinquency data publishing. Quiet week for this signal specifically.”

“The Nvidia/asset-manager deal still has no signed final agreements and no remittance data. But the wider question this signal asks (who ends up holding the paper?) got an official answer. The Bank of England's Financial Policy Committee record (published Sept 30) cites Morgan Stanley estimates: global AI-related debt issuance was about $450B by early September, more than double all of 2025, and about $700B of 2026-28 data-centre capex is expected to be financed by private credit. It warns that 'increasing leverage, opacity and, at times, circular arrangements' could 'amplify losses if expectations disappoint.' Two concrete examples landed the same week. SoftBank's record $11.1B junk bond (up to 9.75%) funded its final OpenAI tranche. And Oracle's force majeure notice on Project Jupiter went to Blue Owl, a private-credit firm. So the lease risk on the Stargate site sits with private credit, exactly the relocation this signal anticipated. Still 'watch'. The risk is relocating faster than any credit tape is being published.”

“No change this week — still MOUs only (Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs, KKR alongside Nvidia), no signed final agreements found, and no remittance or delinquency data publishing. Quiet week for this signal specifically.”

“No change this week — still MOUs only, no signed final agreements found, and no remittance or delinquency data publishing. Quiet week for this signal specifically.”

“No movement this week — H.R.4142 / S.2177 remain in committee with no markup or floor vote found, and no other new Chinese-AI restriction bill or executive action surfaced. Ladder holds at DRAFT_LEGISLATION.”

“Still quiet — no committee markup, floor vote, or new executive action found on H.R.4142 / S.2177 this week. Ladder holds at DRAFT_LEGISLATION.”

“No committee markup, floor vote, or new executive action found this week either. One caveat: Congress.gov's live action log couldn't be directly checked this week (a network restriction blocked the fetch), so this is a search-based 'no news found' rather than a confirmed 'nothing happened' — worth a direct Congress.gov check next sweep. Ladder holds at DRAFT_LEGISLATION.”

“Checked directly against Congress.gov this week: H.R.4142 remains referred to the House Committee on Oversight and Government Reform, and S.2177 remains with Senate Homeland Security and Governmental Affairs — no markup, floor vote, or other action on either bill. Ladder holds at DRAFT_LEGISLATION.”

“No new committee markup, floor vote, or executive action found this week on H.R.4142 or S.2177. Ladder holds at DRAFT_LEGISLATION — the bill text and sponsor list (Krishnamoorthi/Moolenaar in the House, Scott/Peters in the Senate) are unchanged from prior weeks.”

“No committee markup, floor vote, or new executive action found this week on H.R.4142 or S.2177 (a direct Congress.gov check wasn't re-run this sweep, so this is a search-based 'no news found' rather than a confirmed 'nothing happened'). Ladder holds at DRAFT_LEGISLATION.”

“A direct Congress.gov fetch could not be completed this sweep (network restriction), so this is again a search-based 'no news found' rather than a confirmed 'nothing happened.' No markup, floor vote, or new executive action surfaced in general search either. Ladder holds at DRAFT_LEGISLATION.”

“Reconfirmed this week at the same ~40-41% level — no material change. Still record-high concentration, just not moving faster or slower than before.”

“Eased to about 37.6% as of Aug 28, down from ~40-41%. Still within the range this measure has bounced around in recent months and still historically extreme concentration — not read as a material de-concentration of the index.”

“Ticked back up to about 38% as of Sept 1, from 37.6% the week before. This is still just normal week-to-week wobble within the same historically extreme range this measure has held for months — record-high concentration, not a new trend in either direction.”

“No clearly-dated new reading this week — figures found still cluster around 37-38%, the same range logged for the past month. Nvidia alone was cited at roughly 8.2% weight as of Sept 10. Holding at 38% pending a fresher print.”

“Still no cleanly-dated new print for this exact week — figures found range 33-40% depending on whether they measure the 'Magnificent 7' or a strict top-10, with Nvidia cited around 7.5% alone. Consistent with, not clearly beyond, the ~38% already logged. Holding at 38% pending a fresher, methodology-consistent print.”

“Multiple commentary pieces this week (early September dated, still the freshest available) put the top-10 weight at roughly 38-40% of the index — 'ten stocks are now 38% of the S&P 500' is the most commonly cited figure, alongside a Wealth of Common Sense piece framing this as historically extreme versus even the dot-com peak (~26% in 2000) or the 1960s Nifty Fifty (~30% in 1963). No material move found this week either direction. Holding at 38%.”

“First reading from S&P Dow Jones Indices' own data rather than derivative commentary: as of Sept 30, 2026 the top 10 constituents were 39.1% of the index (largest single name 8.4%), across 503 members worth about $68.4T combined. That's a slight tick up from the 38% logged Sept 1 — still inside the same 37-40% band this measure has held for two months, not a new trend in either direction.”

“First read since the Sept 30 S&P Dow Jones Indices snapshot: a Slickcharts aggregation (Oct 10) puts the top-10 weight at 39.03% — Nvidia 7.78%, Apple 6.91%, Microsoft 5.59% — essentially flat versus the official 39.1% logged three weeks ago. This is a derivative/aggregator reading, not the index provider's own data, but it's consistent: no material move in either direction this week.”

“No fresher official print found this week; the Oct 10 Slickcharts reading (39.03%) remains the latest. A separate Oct 2 SPDR S&P 500 ETF holdings snapshot shows Nvidia+Apple+Microsoft alone at 21.59% of the fund, consistent with — not materially beyond — the level already on file. Holding at ~39%.”

“First concrete auction data point for this signal: the Aug 13 30-year bond auction ($31.3B) drew a bid-to-cover of 2.39 — weaker than the trailing 12-month average of 2.43 — at a high yield of 5.216%, the highest 30-year yield in 25 years. Demand wasn't a failure, but it was softer than normal at a moment when the government needs buyers to keep showing up. Worth watching whether this is noise or the start of a trend.”

“The exact mechanism this signal watches for showed up live: a weak 20-year Treasury auction in mid-August briefly pushed 30-year yields above 5.08% and triggered a sharp same-day selloff in AI-adjacent stocks (CoreWeave -8.3%, Tempus AI -6.9%) before the Treasury's doubled bond buybacks calmed the market back down. Then, on Aug 28, incoming Fed Chair Kevin Warsh used his first Jackson Hole speech to say rate hikes may still be needed given elevated inflation — a hawkish surprise that pushed 2-year yields from 4.22% to 4.30% and complicates the idea that buybacks alone can keep the government's own borrowing costs pinned down. Two genuinely new threads for this signal, both pointing the same direction: rising long-term borrowing costs are now visibly colliding with AI-linked asset prices.”

“A genuinely quiet week for new auction data — no Warsh follow-up speech found since Jackson Hole, and no fresh bid-to-cover print. The next real data points are imminent: Treasury has scheduled a 10-year note reopening for Sept 9 and a 3-year note auction for Sept 10, both announced Sept 3 and settling Sept 15 — those results, plus how AI-linked stocks react around them, are this signal's next checkpoint.”

“The Sept 9 10-year note reopening came in strong, not weak: bid-to-cover of 2.71 (the best of the last 28 nominal 10-year auctions since June 2024) at a yield of 4.834%, with indirect (largely foreign) bidders taking 79.2% versus a 72.7% average — real demand for US debt showed up when it mattered. The Sept 10 3-year note auction result could not be confirmed this week. Despite the strong auction, the 10-year yield itself kept climbing, reaching roughly 4.97-4.98% by Sept 11 — the highest since October 2023 — after a hotter-than-expected inflation print pushed the market's odds of a Sept 15-16 Fed rate hike up sharply. Read together: buyers are showing up fine, but the price they're demanding (the yield) keeps rising anyway, which is the more important half of this signal's story than the auction mechanics.”

“The Fed actually delivered the hike the market had been pricing: a 25bp rise to 3.75-4.00%, voted 12-0 on Sept 16 — the first hike in three years, with Chair Warsh citing persistent inflation. The 10-year yield spiked to about 5.02% on Sept 15 (its highest since 2007) and the 30-year hit 5.29% on Sept 17, before both eased back into the high-4s by Sept 17-18 as a separate ~4%+ drop in oil prices calmed inflation fears. Interestingly, gold and silver actually rallied after the hike rather than falling further — a 'sell the rumor, buy the news' pattern — which suggests markets had already priced much of this hike in before it happened. For this signal specifically, the headline point stands: borrowing costs for the government (and everything priced off the 10-year) are now higher than at any point since 2007, which raises the cost of every dollar of debt this AI buildout depends on.”

“A genuinely weak auction this week, and a contrast to two weeks ago's strong one: the Sept 23 5-year Treasury note auction ($70B) priced at a high yield of 5.033% with a bid-to-cover of just 2.21 — the second-largest 'tail' (gap between the yield the market expected going in and the yield it actually cleared at) on record, and the weakest demand for a 5-year note since 2018. Indirect (largely foreign) bidders took only 54.3%, direct bidders 29.9%. This is the opposite pattern from the Sept 9 10-year reopening, which was one of the strongest in years — a reminder that demand for US debt is not uniformly weak or strong right now, it's genuinely inconsistent auction to auction, which is itself worth watching since this AI buildout's borrowing costs depend on that demand staying reliable across every maturity, not just the ones that happen to go well.”

“CORRECTED Oct 6: the original Oct 4 entry said a US government shutdown began Oct 1 and the 10-year yield eased to ~4.15%. Both facts were from October 2025, not 2026. There was no Oct 1 2026 shutdown: a stopgap bill (H.R. 6500) signed Sept 3 2026 funds the government through Dec 11 (Ballotpedia, Sept 25 2026). The 10-year yield did the opposite of easing. It closed between 5.24% and 5.29% every day of the week of Sept 28 (Fed H.15). No new auction result that week. The next nominal auctions (3-year Oct 6, 10-year Oct 7, 30-year Oct 8) fall in the coming week.”

“Borrowing costs have kept climbing, not easing. The Fed's H.15 data puts the 10-year yield at 5.24-5.29% every day from Sept 28 to Oct 2 (5.29% on Sept 30, 5.28% on Oct 2), up from the 5.02% Sept 15 level that was already the highest since 2007. Intraday it touched about 5.31% on Oct 1, the highest since 2002. The 30-year closed at 5.63% on Oct 2. Even a weak September jobs report (Oct 2: payrolls +29,000 vs ~84,000 expected, unemployment up to 4.2%) only briefly pulled yields lower before they closed higher on the day. Separately, the Bank of England's Financial Policy Committee (record published Sept 30) warned that if AI productivity hopes are reassessed, the damage could reach 'not only AI-related asset valuations but also sovereign debt markets.' That is this signal's thesis, stated by a central bank. Status stays 'watch'. The Oct 6-8 3/10/30-year auctions are this week's test.”

“This signal got its first live stress test since this tracker started. An Oct 14-15 AI-led equity selloff (Nasdaq -1.9% on Oct 14, its worst day since July; Philadelphia Semiconductor Index -4.3%, a sixth straight losing session) followed a hotter-than-expected September CPI (headline 3.7% vs 3.6% expected, core 2.6%) and growing unease about AI-linked debt. The 10-year yield fell roughly 8-10bp to about 5.09-5.11%, a three-week low — money moved INTO bonds during the scare, the 2008 pattern (fear points at the stock, not the government), not the 'scared money walks from Treasuries too' alternative this signal exists to catch. BlackRock Investment Institute cut its US equity view to neutral from overweight, citing AI-driven volatility. A reassuring read for now, but one data point — the next real test would be a selloff severe enough to also hit a Treasury auction.”

“The selloff extended through the rest of that week (the S&P 500 and Nasdaq posted their biggest weekly drops since late June) before AI stocks rebounded Oct 19 (Nasdaq +1.11%) as investors came back in; the semiconductor index snapped an 8-day losing streak, its longest since 2010. Specific Oct auction results (3/10/30-year, held Oct 6-8) could not be confirmed this week from a reliable primary source — treat any 10Y level above 5.3% circulating online with caution, since some appears to recycle early-October figures. The next real test lands Oct 28: the Fed's rate decision (hold expected, 3.75-4.00%, ~80% market odds) falls in the same week as Big Tech's Q3 earnings. Status holds 'watch'.”