NVIDIA Printed $96 Billion. Warsh Warned on 3.7%. The Stress Sits in the Chip-Making Machines.

ByEvangelos Papadopoulos

Is compute the new economy? The first published Atlas of the 99-name tech manifold answers in geometry, not adjectives: the market now prices compute as core infrastructure — and prices the factories that build it as the risk.

Three Manifolds · Tech Manifold Special · Issue #18 · Sunday 30 August 2026 · Reading as of Friday 28 August 2026 close · 10 min read · Updated 5 September 2026 — see correction note below

Correction, 4 September 2026. The version published on 30 August gave Nvidia’s dispersion score “by Friday’s close” as +1.7%. That figure was Thursday 27 August’s close: the platform’s Friday run had not ingested Friday’s US closes, a data-calendar defect we identified and corrected on 2 September. On the corrected calendar, Nvidia’s Friday residual is +0.1%. The passages concerned — the opening, the three numbers, the section “where the print landed” and the dollar decomposition — have been restated below; the panel-wide mean absolute dispersion moves from 3.3% to 3.5%, and every other reading in this issue is unchanged. The direction of the finding is not affected, but a number was wrong and we would rather say so than let it stand.

If your book carries the AI complex through the megacap platforms, your risk is not where your hedge is. Measured on the correlation geometry of ninety-nine technology-linked names, the structural concentration this week does not sit in Microsoft, Alphabet or the buyers of compute at all. It sits three layers down — in semiconductor equipment, chip design and the security software layer — and it does not, on the current reading, run through the Fed.

Wednesday evening, Nvidia reported revenue of $96.2 billion for its fiscal second quarter, up 106% from a year ago, with data-centre revenue of $89.0 billion and a guide implying roughly 70% revenue growth in fiscal 2028 — against a Wall Street consensus of 44%. Jensen Huang compressed the thesis into five words: “Now, compute is revenue.” Thursday, the stock closed up 8.7% — its first positive earnings reaction in several quarters — and lifted the Nasdaq 1.6%. Friday morning, Kevin Warsh used his first Jackson Hole keynote as Fed Chair to warn that inflation remains too high, hours after July PCE printed 3.7% against a 3.6% consensus. The tape gave part of Thursday back, and the semiconductor complex led the giveback.

The consensus read the week as an AI blow-off tempered by a hawkish Fed — momentum against monetary gravity. The manifold read something quieter and stranger. By Thursday’s close, the market had repriced Nvidia’s $96 billion quarter to within 1.7% of the structural anchor implied by its own peer geometry — and by Friday’s close, after the giveback, to within 0.1% — and the one block of the tech system that would carry a Fed story, the fourteen rates-and-currency instruments we hold inside the panel precisely to catch macro transmission, was measurably less coherent than at almost any point in its recorded history. Whatever moved through technology this August, it did not arrive through the rates channel.

Three numbers for Monday’s book review:

  • NVDA σ +0.1% at Friday’s close, suture regime (+1.7% at Thursday’s) — a $96 billion quarter absorbed to structural equilibrium within two sessions. In the contagion network, Nvidia’s role this week is absorption, not emission: it soaks up stress; it does not transmit it.
  • Semiconductor-equipment block TSS* 26.2 — the most concentrated of the nine blocks of the tech system, in the fourteenth percentile of its own history, and the site of the most recent block-level curvature alarm, fired 20 August — six days before Nvidia reported.
  • Macro-context block TSS* 47.1 — the 99.8th percentile of its own diversification history, carrying 0% of the system’s alarm attribution. On the week the Fed Chair warned on inflation, the rates block transmits nothing coherent into tech.

Below, the first published Atlas of the ninety-nine-name tech manifold — nine blocks, one meta-manifold, where August’s four block alarms actually fired — and, first, the result of the test we posted publicly before Wednesday’s print, exactly as promised. The operational point stands in one sentence: the market is pricing compute as the economy’s core infrastructure, and pricing the supply chain that manufactures it as the residual risk — and those are two different books to hedge.

Book a discovery session on your own technology exposure →

On the FCI, TSS, σ, regime and epicenter notation → The measures below are experimental research metrics proposed by the author in the Universal Risk Framework (SSRN working papers) — not established market standards such as value-at-risk — and are defined here so the reading is self-contained. TSS is the Topological Survival Score (URF-2, SSRN 6140809): the ratio of the geometric to the arithmetic mean of the panel’s correlation spectrum, in percent — 100% when risk is spread evenly across every direction of the structure, falling toward zero as it concentrates onto a few. Zones — Singularity, Tension, Isotropic, with canonical thresholds 50 and 80 applied to the dimension-normalised TSS* — carry the qualitative reading. FCI is the Flash Condition Index (0–1), capturing how tightly the panel co-moves as a single stress cluster — readings above 0.55 signal an integrated stress regime, readings below 0.35 an idiosyncratic one. σ is the dispersion score: how far a single asset’s spot price sits from its geometric-equilibrium (geodesic) price on the curved manifold of correlation matrices — the value it would carry if it were perfectly aligned with the historical risk-premium structure of its cross-asset peers. An implied structural reference, not a forecast or a target. The regime label — rupture (spot stretched clearly beyond the geodesic), suture (spot within a tolerance band around it) — characterises the geometric state. Contagion zones describe topological position: CORE absorbs the most contagion-mediated stress, induced carries directional emission, periphery sits at the structure’s edge. The epicenter is the network’s dominant source of emitted stress — the emitting node with the strongest combination of contagion outflow and stress intensity. Full methodology at econosysmographe.com/methodology.

New this issue — the Atlas. This is the first publication built on the two-tier hierarchical read of the Tech Manifold. The ninety-nine names are organised into nine economically coherent blocks — semiconductor equipment, chip design, US platforms, software & data, AI infrastructure, cybersecurity, Asia tech, Europe tech, and a macro-context block of rates, currencies and commodities — each analysed as its own manifold, with a meta-manifold measured across the blocks. Each block carries a TSS* score — the same survival score defined above, computed within the block: high means diversified — many independent directions of movement; low means concentrated — the block increasingly moves as one trade. The block topology was frozen on 26 August under a versioned registry derived from our canonical asset dictionary; it is never edited by hand. One consequence we flag openly: because the panel was extended from 52 to 99 names under the new registry, week-on-week comparisons of panel-level scores are suspended for this issue. Every reading below is a measured state as of Friday’s close, set against each block’s own multi-year history — not against last week’s smaller panel.

Two re-costings before Monday nine → One: protection bought on the megacap platforms is insurance on the quiet part of the structure — the platforms read diversified with zero alarm attribution, while the concentrated tier sits in equipment, design and security, where a single-name shock propagates block-wide. A book paying for the first while carrying the second unhedged is paying premium on the wrong building. Two: macro protection bought against the September Fed date is priced against a channel that is measurably not transmitting — the rates block inside the panel is at the 99.8th percentile of its own diversification history and currently organises nothing in technology. The cost that matters is not the hedge that fails; it is the one that was never needed while the supply-chain concentration went uncovered. Full three-point reading in the allocator section below.

Atlas of the Tech Manifold, 28 August 2026: nine blocks with TSS* scores — semi equipment 26.2, semis design 37.8, cybersecurity 38.7, macro context 47.1 in the stressed zone; asia tech 74.6, europe tech 70.9, US platforms 70.8, software & data 65.9, AI infra 63.9 diversified. System TSS* 61.1.
The first published Atlas of the Tech Manifold Live SPD(99), Friday 28 August 2026 close. Tile shade = TSS* zone (low = concentrated). The stressed tier is the compute supply chain — equipment, design, security — not the platforms that buy the compute.

First, the audit

Two sets of commitments come due this issue: a test we posted publicly on Wednesday morning, before Nvidia reported, and three observations pre-registered at Issue #14, our previous Tech special. We publish the audit before the analysis, as always — including the misses.

The Wednesday test. Hours before the print, we posted the measurement that the market had stopped trading Nvidia as a company: the part of its movement belonging to the stock alone carried less than 1% of the 99-name panel’s variance, while the chip-equipment block read as the most compressed structure on the board. The test was written down before the event: if the report surprises, Nvidia’s own direction must reflate; if it does not, the compression stands. The report surprised — a $4.3 billion revenue beat, a guide above the Street, the stock up 8.7% on Thursday. Here is the promised measurement, Tuesday’s close against Friday’s, on the same 99-name panel:

Measured Tue 25 Aug close (as posted) Fri 28 Aug close Read
Nvidia’s own direction — share of panel variance 0.97% 0.27% did not reflate — shrank further
Nvidia’s own direction — systemic norm 0.077 0.10 at the structure’s edge, both sides of the print
Chip-equipment block TSS* 25.6 26.2 compression stands
Chip-design block TSS* 35.1 37.8 compression stands
System TSS* (meta) 57.1 61.1 diversification rose on the print

Verdict: the compression stands — and deepened. The stock rose 8.7% while its own factor fell to roughly a quarter of its already sub-1% variance share: the post-print move was carried by the shared factors of the compute complex, not by an Nvidia-specific direction. The market repriced the complex, not the company — which is the geometric restatement of Huang’s own thesis. (Small print, stated rather than smoothed: the block figures posted Wednesday came from the 26 August snapshot; recomputed under Friday’s snapshot the same readings differ by at most a point — equipment 26.3, design 36.2, system 56.8 — with direction and ranking unchanged.)

The Issue #14 observations. This issue arrives one week earlier than the pre-registered date; anything still open will be re-audited at Issue #19.

  • Observation 1 (does Nvidia’s Q2 FY27 disclose or imply hyperscaler concentration above 50% of data-centre revenue?): NOT YET AUDITABLE. The earnings release and call did not disclose customer concentration; the quarterly 10-Q, where the two-customer disclosure has historically appeared (39% of total revenue combined a year ago), was not yet parsed at the time of writing. Full audit at Issue #19. What the call did disclose points the same direction: an expanded AWS partnership deploying two million Nvidia GPUs across its global network.
  • Observation 2 (does Meta revise its 2026 capex range downward at Q3 earnings?): NOT YET OBSERVABLE — Meta reports in late October. Carried open to the autumn issues.
  • Observation 3 (does any hyperscaler announce new nuclear PPA capacity above 1 GW during August 2026?): NOT TRIGGERED on the public record as of Friday’s close. The nearest event is a reported 1.2 GW hyperscaler power deal in Texas described as near-final, not announced. The “ordering window closing” thesis neither gains nor loses a data point this month.

Zero for three triggered on the carried set; one for one measured on the Wednesday test. We publish both scorelines without adjustment, because the alternative — reporting only the observations that land — is how frameworks stop being falsifiable.

The week: a $96 billion print, a five-word thesis, and a Fed Chair’s first warning

The numbers Nvidia reported Wednesday were large even against the expectations built on them. Revenue of $96.22 billion against a $91.90 billion consensus, up 106% year-on-year. Data-centre revenue of $89.02 billion, up 117%. Adjusted gross margin of 75%, up 250 basis points from a year ago. Adjusted earnings of $2.22 per share against $2.08 expected. Guidance of $108 billion, plus or minus 2%, for the current quarter, against a $105.16 billion consensus — and, on the call, a management expectation of roughly 70% revenue growth in fiscal 2028, far above the 44% analysts had modelled. Huang’s framing was categorical: AI has reached “its inflection point” — and then the five words that give this issue its question: “Now, compute is revenue.”

The tape’s verdict came in two parts. Thursday: Nvidia up 8.7%, the first positive post-earnings session in several quarters, Nasdaq up 1.6%, the European technology complex up 1.8% in early trading. Friday: Warsh’s first Jackson Hole keynote as Chair — the 2% PCE goal a “firm, fixed target”, recent readings “better than expected” but not evidence that “underlying trends have meaningfully improved” — delivered hours after July PCE printed 3.7% year-on-year, a tenth above consensus. The S&P 500 slipped 0.25% to 7,711.76; the Nasdaq lost 0.52%, and semiconductors led the decline. For the week: S&P 500 +0.5%, Nasdaq +0.9%, with only three of eleven S&P sectors higher.

Read as narrative, the week is a tug-of-war: AI momentum against monetary gravity. Read on the manifold, the two forces are not even operating in the same part of the structure.

Start with where the print landed. At Thursday’s close, the session of the 8.7% jump, Nvidia’s dispersion score stood at σ +1.7% — spot price within two percent of its geodesic anchor, comfortably inside the suture band. By Friday’s close, after the tape had given part of Thursday back, it stood at σ +0.1% — spot and anchor all but coincident. For scale: the entire ninety-nine-name panel closed the week inside its suture band — zero names in rupture, a mean absolute dispersion of 3.5%, with the single widest gap in the whole panel a US collaboration-software name at +12.8%, still inside the band. A quarter that came in more than $4 billion above consensus, with a guide nearly $3 billion above the Street, moved that company’s structural position by almost nothing: the market absorbed the number into the existing peer geometry within two sessions. This is what it looks like, geometrically, when a result — however large — confirms the structure the market has already priced rather than breaking it.

Now put dollars on that decomposition, because separating the speculative from the structural is what the instrument is for. Thursday’s 8.7% added roughly $450 billion of paper market capitalisation to Nvidia in a single session, on a base above $5 trillion. At Thursday’s close, the remaining gap between Nvidia’s price and its structural anchor — the part of the price the peer geometry does not yet account for — was σ +1.7%: on the order of $90 billion, about a fifth of what the session added; the other four-fifths the structure absorbed as it repriced the compute complex around the print. By Friday’s close, after the giveback, the residual was σ +0.1%: on the order of $5 billion, about one percent of Thursday’s add. That is the split between the speculative and the geometric component of the week’s marquee move, measured rather than argued: the tape’s enthusiasm, net of structure, had shrunk within two sessions from a roughly $90 billion residual to a rounding error on a five-trillion-dollar name — the structure metabolised it almost entirely, and part of the metabolising happened through price on Friday. The usual epistemics apply: the anchor is an implied structural reference, not a target, and a residual can resolve through price or through structure.

The system-level read agrees. The meta-manifold’s diversification score — TSS* measured across the nine blocks — jumped +4.4 points on Friday’s session alone, from 56.7 to 61.1, its strongest one-day gain of the summer and the 84th percentile of its 437-day history. High TSS* means many independent directions: the blocks de-synchronised on the news pair, they did not converge on it. On the FCI the same signature: 0.29, firmly in the idiosyncratic zone. There is, on this reading, no integrated stress regime in technology this week. Neither the most-watched earnings print of the year nor the most closely parsed Fed communication of the summer created one.

That is the first anti-consensus finding. The second is sharper.

The Atlas: where the system actually carries its concentration

The Atlas exists because a ninety-nine-name correlation panel read flat is a wall of numbers, and read hierarchically is a map. Nine blocks, each its own manifold; one meta-manifold across them; and an attribution module that assigns each block its share of the system’s recent curvature alarms. Here is the full first read, Friday’s close:

Block Names TSS* Reading Share of recent alarm attribution
Semiconductor equipment 9 26.2 concentrated — 14th percentile of own history · block alarm 20 Aug 9.1%
Chip design (Nvidia’s block) 14 37.8 concentrated — 17th percentile 10.1%
Cybersecurity 9 38.7 concentrated — percentile 1.4: historic low · block alarm 5 Aug 20.7%
Macro context (rates, FX, commodities) 14 47.1 99.8th percentile of own diversification history 0.0%
AI infrastructure 9 63.9 diversified 0.0%
Software & data 10 65.9 diversified 0.0%
US platforms 13 70.8 diversified 0.0%
Europe tech 10 70.9 diversified · block alarm 17 Aug 32.9%
Asia tech 11 74.6 diversified · block alarm 13 Aug 27.2%
System (meta, across 9 blocks) 99 61.1 84th percentile · +4.4 pts on Friday’s session HHI 0.24 (uniform 0.11)

Three structural facts stand out, and none of them is visible in an index.

First: the stressed tier of the tech system is the compute supply chain, not the compute buyers. The three blocks reading concentrated — equipment at 26.2, design at 37.8, security at 38.7 — are the layer that manufactures, designs and secures the infrastructure. The blocks that buy the compute — US platforms at 70.8, software & data at 65.9, AI infrastructure at 63.9 — read diversified, quiet, and carry a combined zero percent of the system’s alarm attribution. The cybersecurity reading deserves its own sentence: in the bottom two percent of its own multi-year history (percentile 1.4), the nine-name security block has never traded as a single crowded direction to this degree in our records. When a block’s internal dimensionality collapses toward one, whatever hits one name propagates to all of them — that is the precise sense in which a low TSS* is a fragility reading and not a valuation opinion.

Second: August’s alarms walked through the supply chain before the print. Four block-level curvature alarms fired in the three weeks before Nvidia reported: cybersecurity on 5 August, Asia tech on 13 August, Europe tech on 17 August, semiconductor equipment on 20 August. Over the trailing attribution window, three blocks — Europe tech, Asia tech, cybersecurity — account for 81% of the system’s normalised alarm share, against a concentration index of 0.24 where 0.11 would be uniform. We state the epistemic position plainly, as always: these alarms are measured conditioning, not forecasts. They tell you where the structure was already tightening while the tape was watching a single earnings date. They do not tell you what Monday brings.

Third: the Fed channel is, measurably, not the transmission channel. The macro-context block exists inside the tech manifold for exactly one reason — to catch the moment when rates, currencies and commodities begin to steer technology as a bloc, as they did in 2022. This week, that block sits at the 99.8th percentile of its own diversification history: the fourteen macro instruments are moving with less common structure than at almost any close in the record, and the block carries 0% of alarm attribution. One precision, so the two scales are not misread as a contradiction: at TSS* 47.1 the block still sits below the absolute 50 mark — concentrated in absolute terms, and tinted warm on the Atlas map — while standing at the very top of its own multi-year range. The colour answers “how concentrated now”; the percentile answers “compared with its own past”. Both are printed, deliberately. On the very week the Fed Chair issued his firmest inflation warning yet, the geometry finds no coherent macro-stress direction entering technology. The consensus frames September’s FOMC as the tech trade’s next existential date. The manifold’s finding is narrower and more useful: as of Friday’s close, technology’s internal stress lives in its own supply chain, and the rates complex — whatever it does to the discount rate — is not currently organising the sector’s co-movement. If that changes, this block will say so first, and we will print it.

The epicenter is in Beijing, and Nvidia is absorbing, not emitting

One more layer down, the contagion network across all ninety-nine names names its own protagonist, and it is not the one on the front pages. The week’s epicenter — the network’s dominant source of emitted stress — is Baidu (core zone, emission, systemic norm 0.197), sitting 4.6% below its geodesic anchor. The names it loads against most strongly are not Chinese: a cluster of US software-monitoring and AI-power-infrastructure names. The pattern the contagion topology exposes here is the one no aggregate reads: stress radiating from a Chinese AI platform priced below structural equilibrium into the US names that monitor and power the buildout.

Around it, the semiconductor chain’s roles are almost a diagram of the Atlas finding. TSM sits in the core as an absorber — the node through which stress pools, systemic norm near zero, parked almost exactly on its geodesic. SK Hynix, out on the periphery, is one of the network’s heaviest emitters while sitting 2.2% below its own anchor; Samsung emits from the core at −1.2%. The entire memory complex, in other words, closed the week on its structural anchor — at the same moment the sell side began asking whether memory capacity is the binding constraint on Nvidia’s 70% guide, with one Citrini analyst observing that the number reads “as if NVIDIA has already secured that much memory capacity.” The geometry does not answer the capacity question. It does say the market has not yet priced a memory-scarcity premium into the structure: if the constraint bites, the repricing is still ahead, and it starts from equilibrium.

And Nvidia itself? Systemic norm 0.10 — and its stress direction is absorption. The most-watched name in world markets is not, this week, a transmitter of technology stress. It is a sink for it. Six days after a block alarm fired in the equipment layer that builds its chips, Nvidia printed $96 billion and the network’s read on it barely moved. That pair of facts — supply chain concentrated and alarming, keystone name sutured and absorbing — is this issue’s entire diagnostic, stated twice by two independent measurements.

For an allocator

The register here is diagnostic, not prescriptive; these are configurations to re-cost, not positions to take. But the purpose of the instrument should be stated in money terms, because that is the only test that matters: this geometry exists to locate where repricing risk is concentrated before it reprices. Concentration readings are avoided-loss information — they name the part of a book where one shock moves everything at once. Equilibrium readings are entry-cost information — they name where a thesis, if you hold one, is not yet paid for in the price. Three applications of that test this week:

  • A hedge on the platforms is premium spent on the quiet part of the structure. The megacap buyers of compute read diversified with zero alarm attribution; the concentrated tier — equipment, design, security — is where a single-name shock propagates block-wide. A book hedged at the top of the stack and long the bottom of it is carrying more basis than the index implies — and paying carry for protection that does not cover the exposure most likely to gap. The avoided loss lives in re-matching the hedge to the concentrated tier; the wasted premium lives in not doing so.
  • The memory complex sits on its anchor while the sell side debates memory scarcity — which means the scarcity thesis is not yet in the price. SK Hynix, Samsung and the memory names closed the week on their structural anchors. If the capacity constraint behind Nvidia’s 70% guide is real, the repricing is still ahead and starts from equilibrium — on this measure, an investor who holds that view is not late. If the constraint is not real, the same reading says there is no crowded premium to give back. That asymmetry — thesis optionality still priced at structural fair value — is precisely what the geodesic anchor is for.
  • Macro protection against the September Fed date is priced against a channel that is not transmitting. The macro block’s historic incoherence is a measurement, not a forever: it says the FOMC arrives into a tech structure that rates are not currently organising — so a rates-shock hedge on tech is, today, insurance against a transmission path the data says is closed. The watch item that would change the economics is not the decision itself but whether the macro-context block’s TSS* breaks downward (concentration returning) in the sessions around it. Observation 4 below puts a dated, printed marker on exactly that.

What we pre-register for Issue #19

Observations, not predictions. They can be checked. They will be reported whether they trigger or not.

  • Observation 1 (carried). Nvidia’s Q2 FY27 10-Q: does the customer-concentration disclosure show two customers at or above 40% of total revenue (a year ago: 39% combined)? We will report the number as filed.
  • Observation 2 (carried). Meta’s Q3 report in late October: any downward revision to the 2026 capex range, after three consecutive upward revisions.
  • Observation 3 (new, geometric). The semiconductor-equipment block closed at TSS* 26.2, the 14th percentile of its own history. We will observe whether it exits its bottom historical quintile (TSS* above 28) by the Issue #19 reading — dimensionality returning to the most concentrated block of the system — or whether the concentration deepens.
  • Observation 4 (new, geometric). The macro-context block carries 0% alarm attribution today. We will observe whether that share remains below 10% through the September FOMC week. If macro attribution wakes up, the “not a Fed story” finding of this issue has a dated, printed falsifier.

Is compute the new economy?

Huang’s five words are a revenue claim. The geometry cannot audit his income statement — but it can say how the market is treating the claim, and the answer this week is: as settled infrastructure. A $96 billion quarter absorbed to within 0.1% of structural equilibrium by the end of the week; a whole ninety-nine-name panel inside its suture band; the compute-buying platforms diversified and quiet; the system’s diversification score rising on the news. Markets do not price revolutions this calmly. They price utilities this calmly.

What the market is not treating as settled is the machinery underneath: the equipment, design and security blocks where dimensionality has collapsed toward single-trade behaviour and where every August alarm fired. If compute is the new economy, then the manifold’s finding is that the market believes it — and has moved the risk premium from the economy’s new core to its new supply chain, exactly where infrastructure risk has always lived. That split — core priced calm, supply chain priced fragile, scarcity thesis not yet paid for — is where this week’s avoided losses and unpaid theses both live. We observe the configuration; we do not predict its resolution. The blocks will report back at the next issue.

If any of the companies named wishes to comment on the reading above, or on the pre-registered observations, correspondence is welcomed at the address below.

Book a discovery session →

Sources

A note on data provenance. All numerical figures cited in this issue — earnings results and guidance, index levels, and every manifold, block and Atlas reading — are drawn from primary public inputs (SEC filings, corporate press releases and earnings-call transcripts, publicly available price data) or computed by our own engine from public market data. No proprietary third-party aggregation is reproduced. Where a news publication is cited for context, attribution is given inline and quoted language is limited to fair-use excerpts. The Atlas block topology was frozen on 26 August 2026 under a versioned registry; because the panel was extended from 52 to 99 names under that registry, no week-on-week panel-level deltas are published in this issue — a limitation we state rather than smooth over.

Disclaimer & Regulatory Status. This document is published for sophisticated, professional, and institutional readers for informational and educational purposes only. It does not constitute investment research within the meaning of regulatory frameworks, nor is it an offer, solicitation, or recommendation to buy or sell any financial instrument. SmartGreenInvest Ltd (Reg. England & Wales No. 14636473) is not an FCA-regulated firm. All analyses are based on public quantitative data and geometric modeling as of the date of publication. Opinions expressed are subject to change without notice. Historical performance and model readings are not indicative of future results.

This analysis is a research reading of publicly reported figures for the trading week ending Friday 28 August 2026, published Sunday 30 August 2026 outside market hours. All materials cited are in the public domain.

SmartGreenInvest Ltd holds no positions in the issuers named in this analysis. The author is a self-directed, long-term buy-and-hold investor in US technology equities, at a scale immaterial to the market capitalisations discussed.

By Evangelos Papadopoulos · Independent Researcher · econosysmographe.com

Similar Posts