{ "_meta": { "publication": "The AI Stack Weekly", "schemaVersion": "2026.05.02", "generatedAt": "2026-09-05T17:44:22.765Z", "canonicalUrl": "https://brianletort.ai/industry/weekly/2026-W36", "markdownUrl": "https://brianletort.ai/industry/weekly/2026-W36/llm.md", "pdfUrl": "https://brianletort.ai/downloads/ai-stack-weekly-2026-W36.pdf", "sourceFile": "src/data/industry/weekly/2026-W36.ts" }, "issue": { "slug": "2026-W36", "isoYear": 2026, "isoWeek": 36, "issueNumber": 20, "publishedAt": "2026-09-05", "art": { "hero": { "src": "/images/industry/weekly/2026-W36/hero.png", "alt": "Abstract editorial hero: three interlocking rings of cyan light — software, silicon, and network — turning as one flywheel on a dark field." } }, "executiveSummary": { "keyTakeaways": [ "ARC Prize's matched-effort Astra comparison shows a 35.9-point runtime-associated spread; version the model, memory policy, context compression, and harness together.", "The week's infrastructure pattern is capital committed to dated future service, not operating capacity: a power-purchase agreement starts in 2028, a 150-terabit-per-second cable targets Q4 2028, and delayed-draw funding releases against milestones.", "Vertiv put 44.2% of UtilityInnovation Group's maximum $2.60B consideration behind EBITDA delivery; interpreting that as seller risk transfer is a house reading, not a disclosed fact.", "Broadcom's $16.7B AI-semiconductor quarter proves supplier monetization, not utilization or application returns; custom accelerators and networking remain combined.", "PJM's comment window closed while Texas staff proposed $50,000/MW security above 75 MW, keeping power obligations inside operating contracts." ], "byTheNumbers": [ { "value": "35.9 points", "label": "Astra matched-effort harness-associated spread", "context": "ARC Prize: 62.7% Standard versus 98.6% Provider Adapter, both at maximum reasoning." }, { "value": "44.2%", "label": "Vertiv-UIG maximum price that is contingent", "context": "$1.15B earnout divided by $2.60B maximum consideration." }, { "value": "$16.7B", "label": "Broadcom Q3 AI-semiconductor revenue", "context": "Filed unaudited quarterly; 56.4% of $29.6B total revenue." }, { "value": "396 MW", "label": "Firm geothermal capacity under Google-Fervo power-purchase agreement", "context": "Four 99 MW tranches beginning Q3 2028; conditional expansion excluded." }, { "value": "$3.05B", "label": "Nscale delayed-draw commitments", "context": "$1.85B Texas plus $1.20B North Carolina maximum commitments." } ] }, "bigStory": { "headline": "AI's scarce asset is now a governed path to service, not an announced model or megawatt", "body": "Astra's benchmark split and this week's infrastructure contracts point to the same operating reality: the asset being bought is a controlled path from intent to completed work, not a standalone model, chip, cable, or megawatt. ARC Prize's matched-effort comparison associates a 35.9-point runtime spread with the harness; Model Pulse carries the Standard and Provider Adapter scores. Infrastructure buyers likewise committed to dated future power, fiber, and deployment paths rather than operating supply.\n\nBoards should require one schedule-and-control ledger across the stack: model plus harness and memory policy, power and fiber service dates, funding milestones, authority scopes, evidence custody, and rollback. Do not count options as contracted capacity, commitments as funded cash, previews as production services, or provider-adapter scores as model-only capability.\n\nVertiv's UtilityInnovation Group structure makes the underwriting visible: 44.2% of maximum consideration remains contingent on undisclosed post-close EBITDA targets. The arithmetic is house-original; the view that this transfers delivery risk to sellers is an interpretation because seller obligations and payout curves are undisclosed.", "arc": "all-three", "keyPoints": [ "Credit ARC Prize's finding and use its matched-effort comparison: Astra's Provider Adapter is associated with a 35.9-point gain at maximum reasoning.", "Separate operating supply from dated future-service commitments across geothermal power, subsea fiber, and milestone-based deployment finance.", "Use milestone, option, earnout, and rollback terms to price delivery risk rather than quoting maximum headline capacity.", "Vertiv left a material share of UIG's maximum consideration contingent on post-close EBITDA delivery; see the house measurement." ], "pullQuote": "The asset being bought is a controlled path from intent to completed work, not a standalone model, chip, cable, or megawatt." }, "lenses": { "software": { "events": [ { "date": "Sep 3", "title": "OpenAI begins phased GPT-6 Astra rollout with 1M context and $10/$50 per million-token Standard API pricing", "source": "OpenAI, ARC Prize, Artificial Analysis", "sourceUrl": "https://openai.com/index/gpt-6-astra/" }, { "date": "Sep 1", "title": "Claude Fable 5.1 becomes generally available across GitHub Copilot; enterprise administrators must enable it", "source": "GitHub", "sourceUrl": "https://github.blog/changelog/2026-09-01-claude-fable-5-1-generally-available-in-github-copilot/" }, { "date": "Sep 3", "title": "Gemini 3.8 Flash enters GitHub Copilot under introductory provider pricing through year-end", "source": "GitHub", "sourceUrl": "https://github.blog/changelog/2026-09-03-gemini-3-8-flash-is-now-available-in-github-copilot/" }, { "date": "Sep 1", "title": "NVIDIA releases Apache-2.0 Muse-Glimmer-30B NVFP4 (4-bit floating-point) checkpoint, reducing storage from 60 GB to 24.7 GB", "source": "NVIDIA on Hugging Face", "sourceUrl": "https://huggingface.co/nvidia/Muse-Glimmer-30B-NVFP4" } ], "takeaways": [ "Version the model, harness, memory policy, and retry budget together; Astra's headline result is not model-only.", "Treat Fable and Gemini availability, retention, administrator enablement, and promotional pricing as stock-keeping unit (SKU) properties.", "See Model Pulse for tree additions, benchmark comparability, and the quiet open-weight week." ], "meaning": "Architects should stop procuring model names as if the surrounding runtime were interchangeable. Astra's neutral-versus-provider harness spread, GitHub's administrator gates, and year-end Gemini pricing all make deployment policy part of measured capability and cost. See The Model Pulse for the full model and tree read." }, "hardware": { "events": [ { "date": "Aug 31", "title": "AMD, Cisco, and HUMAIN put an MI355X and 800G Ethernet cluster into production; live scale and utilization undisclosed", "source": "AMD, Cisco", "sourceUrl": "https://newsroom.amd.com/news/amd-cisco-humain-expand-saudi-arabia-ai-infrastructure/" }, { "date": "Aug 31", "title": "MediaTek joins NVIDIA NVLink Fusion for custom cloud accelerators (XPUs) without a named customer, tape-out, or shipment date", "source": "NVIDIA", "sourceUrl": "https://nvidianews.nvidia.com/news/nvidia-and-mediatek-deepen-long-standing-partnership-to-build-ai-edge-to-cloud-computing-platforms" }, { "date": "Sep 1", "title": "Cerebras contracts phased 50 MW, 80 MW, and 165 MW Finnish campus capacity under seven-year service orders", "source": "Cerebras announcement mirror, Converge Digest", "sourceUrl": "https://convergedigest.com/cerebras-165mw-finland-ai-data-center/" }, { "date": "Sep 2", "title": "Broadcom reports AI-semiconductor revenue at 56.4% of Q3 company revenue, without splitting accelerators from networking", "source": "Broadcom earnings release", "sourceUrl": "https://www.prnewswire.com/news-releases/broadcom-inc-announces-third-quarter-fiscal-year-2026-financial-results-and-quarterly-dividend-302868129.html" } ], "takeaways": [ "Count HUMAIN as production evidence, but not as proof of the 250 MW or 1 GW roadmap ceilings.", "Keep custom-accelerator and NVLink dependence in the same architecture model; MediaTek disclosed an ecosystem commitment, not shipping silicon.", "Broadcom's AI line proves supplier revenue conversion while withholding the split needed to rank compute versus networking economics." ], "meaning": "Infrastructure buyers should separate production systems, contracted phases, and architecture announcements. The HUMAIN cluster is live but unquantified; Cerebras has only its first Finnish phase under construction; MediaTek-NVLink has no delivery milestone. Broadcom's filed unaudited quarterly disclosure is the hard signal, but its combined custom-accelerator-and-networking line cannot resolve which layer holds margin." }, "networking": { "events": [ { "date": "Aug 31", "title": "AWS-Azure managed private multicloud interconnect enters preview in four paired regions", "source": "AWS", "sourceUrl": "https://aws.amazon.com/about-aws/whats-new/2026/08/aws-announces-AWS-interconnect-multicloud-microsoft-azure-preview/" }, { "date": "Sep 1", "title": "DE-CIX deploys dual-homed Azure ExpressRoute Metro across five global metros", "source": "DE-CIX", "sourceUrl": "https://www.de-cix.net/en/about-de-cix/media/press-releases/de-cix-expands-global-ai-ready-connectivity-with-microsoft-azure-expressroute-metro" }, { "date": "Sep 2", "title": "Equinix announces pre-beta Fabric One for intent-driven routing, encryption, resilience, and failover", "source": "Equinix", "sourceUrl": "https://newsroom.equinix.com/2026-09-02-Equinix-Unveils-Equinix-Fabric-One,-Redefining-How-Enterprises-Connect-Across-AI,-Cloud-and-Networking-Infrastructure" }, { "date": "Sep 3", "title": "Firmus commits approximately $300M for up to 150 terabits per second on APX East for 25 years, targeting Q4 2028 service", "source": "Firmus and SUBCO", "sourceUrl": "https://firmus.co/newsroom/firmus-and-subco-announce-major-trans-pacific-ai-connectivity-agreement" } ], "takeaways": [ "Buy route diversity and policy simulation before natural-language network changes reach production.", "Treat Fabric One as architecture direction until pricing, service levels, provisioning time, and rollback evidence publish.", "Firmus bought future schedule position; it did not demonstrate a present bandwidth shortage or lit capacity." ], "meaning": "Network architects should design delegated changes around simulation, approval, evidence, and rollback, not just easier invocation. AWS-Azure and Fabric One make intent-driven multicloud connectivity concrete, while the Firmus contract shows long-duration queue hedging. The durable-layer thesis remains strained until revenue growth proves pricing power." } }, "capitalFlow": [ { "category": "Frontier Labs", "examples": "OpenAI, Anthropic, Google DeepMind, xAI", "capitalIn": "Unknown", "capitalInPrior": "Unknown", "capitalInDirection": "flat", "revenueOut": "Unknown", "revenueOutPrior": "Unknown", "revenueOutDirection": "flat", "burnToRevenue": "n/a", "movement": "No reproducible W36 constituent ledger supports a category point estimate; no in-window financing or annual revenue disclosure establishes one.", "narrative": "Prior rolling estimates are not reproduced because this issue's artifact set lacks the constituent list, valuation date, and formula inputs. Astra pricing and Anthropic's safeguards remain product signals, not category capital or revenue totals.", "transactions": [ { "date": "Sep 3", "label": "OpenAI launches GPT-6 Astra Standard API (list pricing in software events)", "source": "OpenAI", "sourceUrl": "https://openai.com/index/gpt-6-astra/" }, { "date": "Sep 1", "label": "Anthropic announces customer-controlled Enterprise Frontier Safeguards", "source": "Anthropic", "sourceUrl": "https://www.anthropic.com/news/enterprise-frontier-safeguards" } ] }, { "category": "Hyperscaler-Hosted", "examples": "Azure-OpenAI, AWS-Anthropic, Google Cloud-Gemini, Oracle-OCI", "capitalIn": "Unknown", "capitalInPrior": "Unknown", "capitalInDirection": "flat", "revenueOut": "Unknown", "revenueOutPrior": "Unknown", "revenueOutDirection": "flat", "burnToRevenue": "n/a", "movement": "No reproducible W36 constituent ledger supports a category point estimate; Google's Fervo agreement has no disclosed consideration.", "narrative": "Google contracted the firm Fervo geothermal capacity noted above and received an option for approximately 600 MW of conditional expansion. If accepted under a definitive agreement, total contracted capacity would be not less than approximately 950 MW, with June 2030 as the expansion's guaranteed commercial-operation date. None belongs in a capital total without disclosed consideration.", "transactions": [ { "date": "Sep 1", "label": "Google Energy signs 396 MW, 15-year Fervo power-purchase agreement; initial delivery targeted Q3 2028", "source": "Fervo filing mirror", "sourceUrl": "https://www.stocktitan.net/sec-filings/FRVO/8-k-fervo-energy-co-reports-material-event-1de9aacddd08.html" }, { "date": "Aug 31", "label": "AWS-Azure private multicloud interconnect enters four-region preview", "source": "AWS", "sourceUrl": "https://aws.amazon.com/about-aws/whats-new/2026/08/aws-announces-AWS-interconnect-multicloud-microsoft-azure-preview/" } ] }, { "category": "Neoclouds", "examples": "CoreWeave, Nscale, Crusoe, Lambda, Fluidstack, IREN", "capitalIn": "Unknown", "capitalInPrior": "Unknown", "capitalInDirection": "flat", "revenueOut": "Unknown", "revenueOutPrior": "Unknown", "revenueOutDirection": "flat", "burnToRevenue": "n/a", "movement": "Nscale added up to $3.05B of delayed-draw commitments, but the unreproduced carried base prevents an honest category total or direction.", "narrative": "Nscale closed $1.85B for Texas and $1.20B for North Carolina. These are maximum commitments available against future deployment milestones, not cash funded on August 31. They are shown as a transaction, not added to a fabricated rolling ledger.", "transactions": [ { "date": "Aug 31", "label": "Nscale closes two investment-grade delayed-draw facilities", "amount": "Up to $3.05B", "source": "Nscale", "sourceUrl": "https://www.nscale.com/press-releases/nscale-3-billion-delayed-draw-term-loans" }, { "date": "Sep 3", "label": "Firmus reserves future APX East capacity for its Australian AI-factory program", "amount": "~$300M", "source": "Firmus and SUBCO", "sourceUrl": "https://firmus.co/newsroom/firmus-and-subco-announce-major-trans-pacific-ai-connectivity-agreement" } ] }, { "category": "On-Prem / Hybrid", "examples": "Enterprise GPU clusters, sovereign and national programs, Cisco / Dell / HPE", "capitalIn": "Unknown", "capitalInPrior": "Unknown", "capitalInDirection": "flat", "revenueOut": "Unknown", "revenueOutPrior": "Unknown", "revenueOutDirection": "flat", "burnToRevenue": "n/a", "movement": "No reproducible W36 constituent ledger supports a category point estimate; the atNorth deal changes ownership rather than measuring new category capital.", "narrative": "CPP Investments and Equinix completed the atNorth acquisition, and AMD-Cisco-HUMAIN announced a live Saudi cluster. Neither supplies a reproducible aggregate for this category.", "transactions": [ { "date": "Sep 2", "label": "CPP Investments and Equinix complete atNorth acquisition", "amount": "~$4.0B", "source": "Equinix and CPP Investments", "sourceUrl": "https://newsroom.equinix.com/2026-09-02-CPP-Investments-and-Equinix-Complete-atNorth-Acquisition-to-Support-Growth-of-Leading-Nordic-Data-Center-Platform" }, { "date": "Aug 31", "label": "AMD-Cisco-HUMAIN production MI355X and 800G cluster goes live; scale undisclosed", "source": "AMD", "sourceUrl": "https://newsroom.amd.com/news/amd-cisco-humain-expand-saudi-arabia-ai-infrastructure/" } ] } ], "signalVsNoise": [ { "score": 4, "claim": "Broadcom reported Q3 FY2026 revenue of $29.6B and AI-semiconductor revenue of $16.7B, 56.4% of company revenue.", "sources": "Broadcom filed unaudited quarterly earnings release, September 2, 2026.", "read": "The reported line supports supplier monetization. It does not separate custom accelerators from networking or show customer utilization and application returns." }, { "score": 5, "claim": "Google Energy signed a 396 MW, 15-year Fervo power-purchase agreement in four 99 MW tranches beginning in Q3 2028.", "sources": "Fervo filing mirror and Utility Dive, September 1, 2026.", "read": "Count 396 MW as firm future capacity. Exclude conditional expansion until Google accepts it and a definitive agreement is executed; do not treat either amount as energized today." }, { "score": 4, "claim": "Astra scored 99.9% on ARC-AGI-3 and therefore nearly saturates general intelligence.", "sources": "OpenAI and ARC Prize, September 3, 2026.", "read": "The number is real but the conclusion is not. ARC Prize's Standard harness landed far below the headline adapter score; the best-observed adapter result used different reasoning effort, provider-private state, and context compression. Model Pulse carries the matched-effort scores. ARC says the closed-ended benchmark is not proof of general intelligence." }, { "score": 2, "claim": "DeepSeek will deploy at least 160,000 Huawei Ascend accelerators at a roughly 1 GW Inner Mongolia inference site.", "sources": "Bloomberg and The Decoder, September 4, 2026; no primary confirmation.", "read": "Potentially material domestic-inference substitution, but no purchase contract, delivered quantity, rack design, or schedule is public. Do not put it into capacity forecasts yet." }, { "score": 1, "claim": "Fabric One has already made the network an autonomous production control plane for enterprise AI.", "sources": "Equinix announcement and launch coverage, September 2-4, 2026.", "read": "Architecture direction, not operating proof. The product is pre-beta with no price, service level, provisioning benchmark, rollback evidence, or production customer result." } ], "levers": [ { "metric": "Frontier lab cash runway at current burn", "current": "Unknown — W36 has no reproducible cash-and-burn input table", "prior": "Prior estimate withheld for the same reason", "direction": "flat", "threshold": "Below 18 months for any disclosed-burn lab", "detail": "The required inputs are unaudited and absent from this issue's artifact set. Astra pricing and Enterprise Frontier Safeguards cloud-storage costs do not establish a financing input." }, { "metric": "Hyperscaler AI capex to disclosed AI revenue ratio", "current": "Unknown — no reproducible top-four capex and AI-revenue input range", "prior": "Prior point estimate withheld because AI revenue is not segment-audited", "direction": "flat", "threshold": "Above 6x sustained for two consecutive quarters", "detail": "The methodology requires a range and a top-four input table. Neither is present in W36, and the Fervo power-purchase agreement has no disclosed consideration." }, { "metric": "CoreWeave contracted revenue backlog", "current": "$104.2B as of June 30, unchanged — no CoreWeave filing in-window", "prior": "$104.2B as of June 30, unchanged — no CoreWeave filing in-window; Vera Rubin production deep-dive reinforces operational moat ahead of Q3 print", "direction": "flat", "threshold": "Sequential decline, or conversion below 15% annually", "detail": "Backlog remains a filed stock value awaiting the next quarter. Nscale delayed-draw financing is a category signal, not a CoreWeave backlog input." }, { "metric": "NVIDIA quarter-over-quarter data center revenue", "current": "$89.0B for Q2 FY27, held pending Q3; Broadcom separately reported AI-semiconductor revenue (see byTheNumbers)", "prior": "$89.0B for Q2 FY27 (+117% YoY), up sequentially from $75.2B Q1; Vera Rubin production began in August; ~20% datacenter mix guided for Q3 (vendor-stated)", "direction": "flat", "threshold": "Two consecutive quarters of sequential decline", "detail": "No NVIDIA print landed this week. Broadcom's filed unaudited quarterly disclosure broadens supplier evidence but cannot be substituted into NVIDIA's series." }, { "metric": "Open-weight to closed-model capability gap on coding", "current": "Untested this week — Muse-Glimmer NVFP4 cuts checkpoint storage (see software events), but no new independent open-frontier benchmark set", "prior": "Narrowed on vendor-reported rows but substitutability improved: IBM Granite 4.2 30B Apache 2.0 with vendor-reported 57% SWE-Bench Verified; GLM-5.3-Flash MIT weights with vendor-reported 84.3% Terminal-Bench 2.1 — independent tracker confirmation still pending", "direction": "flat", "threshold": "Open weights within 2 Index points of the closed leader", "detail": "The quantized checkpoint improves deployment footprint, not the measured frontier gap. See Model Pulse for the benchmark and lineage read." }, { "metric": "Sovereign AI program commitments", "current": "Unknown — W36 has no reproducible program ledger", "prior": "Prior aggregate withheld for the same reason", "direction": "flat", "threshold": "Above 20 programs or $250B committed", "detail": "The method includes government-funded national compute programs only. Corporate acquisitions, power agreements, and deployments stay outside, but no constituent ledger is available here to support a point estimate." }, { "metric": "PJM capacity auction clearing price", "current": "$325.00 per MW-day for 2028/29, unchanged — ER26-3515 comment deadline passed Sep 3", "prior": "$325.00 per MW-day for 2028/29, unchanged", "direction": "flat", "threshold": "A second consecutive auction clearing at the cap", "detail": "No auction occurred. No capped-versus-uncapped 2028/29 simulation was identified in the evidence reviewed for W36; the policy signal is a pending capacity-or-curtail rule." }, { "metric": "Time from interconnection request to energization", "current": "Unknown — no reproducible multi-queue duration table in W36", "prior": "Prior range withheld for the same reason", "direction": "flat", "threshold": "Below 48 months in two or more major queues", "detail": "W36 makes compliance costs more explicit but does not publish a comparable queue-duration update. At the 75 MW Texas threshold, proposed security equals $3.75M." }, { "metric": "Cost per task, frontier reasoning model", "current": "No commensurate W36 update; ARC Prize suite totals are not a market floor or median", "prior": "W35 named-basis floor: GLM-5.3-Flash vendor-reported $0.045 per task on Artificial Analysis Index 57; no same-basis median available", "direction": "flat", "threshold": "A frontier-tier reasoning model below $1 per million output tokens", "detail": "Methodology v2 requires a named-basis floor and median. ARC Prize's suite totals are retained in Model Pulse but cannot move this lever against the prior Artificial Analysis basis." }, { "metric": "Custom silicon share of hyperscaler AI compute", "current": "Unknown — Broadcom disclosed combined custom-accelerator and networking revenue but no compute-share or component split", "prior": "Unknown — Hot Chips disclosed inference ASIC roadmaps (Google TPU 8i, Jalapeño, Maia 200, MTIA 400) but no in-window hyperscaler compute-mix filing supports a booked share estimate; disclosure cadence ≠ shipped mix", "direction": "flat", "threshold": "Above 45% share with audited hyperscaler mix disclosure", "detail": "The revenue line confirms scale but cannot produce a custom-compute share because networking is included and deployed utilization is absent." } ], "houseMeasurement": { "type": "filing-derived", "headline": "Vertiv made 44.2% of UtilityInnovation Group's maximum $2.60 billion consideration performance-contingent, putting 79.3 cents of earnout behind every upfront dollar.", "method": "House computation from Vertiv's September 2, 2026 acquisition disclosure. The primary release states: (1) approximately $1.45 billion cash consideration at closing; (2) up to $1.15 billion additional cash consideration tied to 12- and 24-month EBITDA targets; and (3) the upfront price is approximately 13 times expected 2027 EBITDA. Arithmetic, all in US dollars: maximum consideration = $1.45B + $1.15B = $2.60B; contingent share of maximum = $1.15B ÷ $2.60B = 44.23%; upfront share = $1.45B ÷ $2.60B = 55.77%; contingent dollars per upfront dollar = $1.15B ÷ $1.45B = $0.793; current implied expected 2027 EBITDA = $1.45B ÷ 13 = approximately $111.54M; maximum earnout ÷ current implied expected EBITDA = $1.15B ÷ $111.54M = 10.31x. The last ratio is a scale comparison, not the deal's full-price EBITDA multiple, because paying the earnout requires undisclosed higher EBITDA targets.", "readings": [ { "label": "Performance-contingent share of maximum consideration", "value": "44.2% ($1.15B of $2.60B)", "context": "$1.15B maximum earnout ÷ ($1.45B upfront + $1.15B maximum earnout); this is the selected house measurement" }, { "label": "Earnout dollars per upfront dollar", "value": "$0.793 per $1.00 upfront", "context": "$1.15B maximum earnout ÷ $1.45B cash at close; the structure places nearly eighty cents of additional price behind each upfront dollar" }, { "label": "Upfront share of maximum consideration", "value": "55.8% ($1.45B of $2.60B)", "context": "$1.45B cash at close ÷ $2.60B maximum; closing cash is still the majority of potential consideration" }, { "label": "Implied expected 2027 EBITDA", "value": "Approximately $111.5M", "context": "$1.45B upfront price ÷ the disclosed approximately 13x expected 2027 EBITDA multiple; approximate because both inputs are rounded" }, { "label": "Maximum earnout relative to current implied EBITDA", "value": "10.31x current implied expected 2027 EBITDA", "context": "$1.15B ÷ $111.5M; this sizes the contingent pool only and is not an acquisition multiple because the earnout depends on higher undisclosed EBITDA targets" } ], "implication": "The disclosed structure makes a material share of potential consideration contingent rather than fixed. The house interpretation is that this transfers some delivery risk to sellers, but the disclosure does not quantify that transfer. Boards should compare the contingent share of maximum consideration and the earnout-to-upfront ratio before comparing headline transaction values.", "caveats": "Seller risk transfer is an interpretation, not a disclosed fact: Vertiv does not publish seller obligations, the EBITDA thresholds, their weighting, payout curve, revenue concentration, or baseline. The approximate inputs limit precision, the $2.60B maximum is not the expected purchase price, and the 10.31x scale comparison is not a transaction multiple. Closing remains expected in Q4 2026.", "sources": [ { "name": "Vertiv acquisition announcement and financial terms", "url": "https://www.vertiv.com/en-us/about/news-and-events/corporate-news/2026/vertiv-announces-agreement-to-acquire-utilityinnovation-group-to-accelerate-time-to-power-for-ai-data-centers/" } ] }, "whereWeDiffer": [ { "theirTake": "GPT-6 Astra's 99.9% ARC-AGI-3 result is near-saturation evidence of broadly human-level agentic intelligence.", "source": "OpenAI and ARC Prize, Sep 3, 2026", "sourceUrl": "https://arcprize.org/blog/astra", "stance": "differ", "ourRead": "The 99.9% adapter result is best-observed and descriptive, but it uses a different reasoning effort from the matched Standard-harness maximum. Model Pulse carries the matched-effort Standard versus Provider Adapter scores. Astra plus its runtime nearly saturates a closed-ended benchmark; model-only saturation and general intelligence do not follow." }, { "theirTake": "Broadcom's 56.4% AI revenue share proves custom-silicon returns have already concentrated upstream.", "source": "Broadcom earnings release; RCR Tech", "sourceUrl": "https://www.prnewswire.com/news-releases/broadcom-inc-announces-third-quarter-fiscal-year-2026-financial-results-and-quarterly-dividend-302868129.html", "stance": "open", "ourRead": "The reported revenue is material, but it comes from a filed unaudited quarterly disclosure, combines custom accelerators and networking, and omits customer utilization, component margins, and downstream returns. It supports shipment monetization, not where durable return pools settle." }, { "theirTake": "Vertiv and Flex spent $7.0B to buy one integrated grid-to-chip control plane.", "source": "Vertiv and Flex announcements, Sep 2-3, 2026", "sourceUrl": "https://www.vertiv.com/en-us/about/news-and-events/corporate-news/2026/vertiv-announces-agreement-to-acquire-utilityinnovation-group-to-accelerate-time-to-power-for-ai-data-centers/", "stance": "differ", "ourRead": "The headline sum collapses different assets. Vertiv buys site architecture and microgrid orchestration with a material contingent share of maximum price (see house measurement); Flex buys conversion hardware and grid-forming controls at 5.5x forecast 2026 revenue. Both are pending acquisitions, not operating capacity." }, { "theirTake": "Equinix Fabric One has turned the network into the autonomous control plane for distributed AI.", "source": "Equinix and SiliconANGLE, Sep 2-4, 2026", "sourceUrl": "https://newsroom.equinix.com/2026-09-02-Equinix-Unveils-Equinix-Fabric-One,-Redefining-How-Enterprises-Connect-Across-AI,-Cloud-and-Networking-Infrastructure", "stance": "extend", "ourRead": "Intent-driven connectivity is the direction, but Fabric One is pre-beta. The control-plane product is not natural-language provisioning alone; it is policy simulation, scoped approval, retained intent history, rollback, and production service evidence that have not yet published." } ], "synthesis": { "connections": [ { "claim": "Enterprise AI is re-layering around a delegated-control boundary: policy, evidence custody, approval, and rollback are becoming separately owned infrastructure around model execution.", "chain": [ "Astra's launch joined persistent model state with action-level classifiers and automatic review, making runtime control part of the released capability rather than an external compliance attachment.", "Anthropic's Enterprise Frontier Safeguards separates detector operation from data custody: monitoring runs across sessions, while logs, keys, flagged-event review, and human access stay in the customer's cloud account.", "GitHub now lets a Copilot approval satisfy branch-protection rules, but scopes the authority by repository and path and dismisses the approval after a new commit, converting AI judgment into revocable policy state.", "Equinix plans to translate portal, application programming interface, agent, and natural-language intent into routing, encryption, resilience, and failover, moving delegated control from software changes into network state." ], "reasoningType": "abductive", "evidence": [ { "label": "agents-01", "sourceUrl": "https://deploymentsafety.openai.com/gpt-6-astra" }, { "label": "agents-02", "sourceUrl": "https://www.anthropic.com/news/enterprise-frontier-safeguards" }, { "label": "agents-03", "sourceUrl": "https://github.blog/changelog/2026-09-01-copilot-code-review-can-now-approve-pull-requests/" }, { "label": "networking-04", "sourceUrl": "https://newsroom.equinix.com/2026-09-02-Equinix-Unveils-Equinix-Fabric-One,-Redefining-How-Enterprises-Connect-Across-AI,-Cloud-and-Networking-Infrastructure" } ], "confidencePct": 76, "steelMan": "Three of the four controls are new, previewed, or rolling out later: Anthropic's Enterprise Frontier Safeguards starts in phases, Copilot approval is public preview, and Fabric One is not yet in beta. These could remain vendor-specific features rather than a durable independent layer. The strongest part of the claim is narrower: across code, model traffic, and networking, vendors are converging on explicit scopes, retained evidence, and revocation. Falsified if production releases fold those controls back into opaque model behavior or customers cannot export and audit the resulting policy state by Q2 2027." }, { "claim": "An accumulating procurement pattern is visible across AI infrastructure: buyers are committing capital to dated future-service paths in power, fiber, and deployment before those assets operate.", "chain": [ "Google's 15-year Fervo power-purchase agreement contracts 396 MW in four tranches beginning in Q3 2028 and includes approximately 600 MW of conditional expansion, reserving a future delivery path rather than consuming present generation.", "Firmus committed approximately $300 million for up to 150 terabits per second over 25 years on an Australia-US cable targeted for service in Q4 2028, paying for route position before the system is built.", "Nscale's $3.05 billion facilities are delayed-draw commitments, meaning funds become available against future milestones rather than all at closing, for two deployments spanning compute, network, cooling, and site work.", "Vertiv and Flex agreed to acquire power-control and conversion capabilities that influence whether a permitted site can move from interconnect application to energized rack." ], "reasoningType": "inductive", "evidence": [ { "label": "capital-02", "sourceUrl": "https://www.stocktitan.net/sec-filings/FRVO/8-k-fervo-energy-co-reports-material-event-1de9aacddd08.html" }, { "label": "networking-05", "sourceUrl": "https://firmus.co/newsroom/firmus-and-subco-announce-major-trans-pacific-ai-connectivity-agreement" }, { "label": "capital-01", "sourceUrl": "https://www.nscale.com/press-releases/nscale-3-billion-delayed-draw-term-loans" }, { "label": "capital-05", "sourceUrl": "https://www.vertiv.com/en-us/about/news-and-events/corporate-news/2026/vertiv-announces-agreement-to-acquire-utilityinnovation-group-to-accelerate-time-to-power-for-ai-data-centers/" } ], "confidencePct": 74, "steelMan": "Power agreements, subsea capacity, delayed-draw loans, and acquisitions are different instruments with different risks; grouping them can obscure ordinary long-lead infrastructure planning. None proves that queue positions are independently tradable or earn excess returns. The inference is a continuing procurement pattern, not a new asset class or measured directional shift. Falsified if comparable power, fiber, and compute capacity becomes repeatedly available on short notice without precommitment or scarcity-linked terms by mid-2027." }, { "claim": "Agent procurement complexity is accumulating in bundles, credits, workflow entitlements, retention routes, and runtime choices, making utilization forecasting a core buying competency.", "chain": [ "Salesforce embedded Agentforce, collaboration, analytics, security, and support into three seat editions while coupling access rights to different consumption pools.", "Astra launched as a premium application-programming-interface tier while Gemini 3.8 Flash entered Copilot under temporary introductory provider pricing, widening the model-price range inside comparable agent surfaces.", "GitHub administrators can choose whether Fable 5.1 is available across nine surfaces, showing that enterprise entitlement and retention policy can matter as much as public model availability.", "AWS's migration pattern deliberately separates tools, loop, runtime, memory, and observability so model choice can change without rebuilding every operating layer." ], "reasoningType": "abductive", "evidence": [ { "label": "applications-01", "sourceUrl": "https://www.salesforce.com/news/stories/salesforce-simplifies-editions-2026/" }, { "label": "software-01", "sourceUrl": "https://openai.com/index/gpt-6-astra/" }, { "label": "software-02", "sourceUrl": "https://github.blog/changelog/2026-09-01-claude-fable-5-1-generally-available-in-github-copilot/" }, { "label": "agents-04", "sourceUrl": "https://aws.amazon.com/blogs/machine-learning/migrate-agentic-workloads-to-amazon-bedrock-agentcore/" } ], "confidencePct": 67, "steelMan": "Salesforce is one vendor, its credits are not directly convertible to model tokens, and introductory Gemini pricing may expire without changing Copilot seat economics. Customers can still capture value primarily from superior capability on difficult tasks. This evidence supports procurement complexity, not margin migration. Falsified if major agent suites converge on transparent per-outcome prices and publish stable task-level usage denominators by Q2 2027." } ], "thesisTest": [ { "hypothesisNumber": 1, "hypothesis": "The cycle is accelerating, not slowing.", "verdict": "strained", "reasoning": "W36 shows rapid software packaging: Astra arrived with persistent memory, asynchronous clarification, and action monitoring, while Anthropic, GitHub, and AWS announced adjacent distribution and control changes in the same week. It does not measure a shorter prior-to-current doubling interval across software, hardware, and networking, and therefore cannot support the hypothesis's flywheel-cadence test.", "evidence": [ { "label": "software-01", "sourceUrl": "https://arcprize.org/blog/astra" }, { "label": "agents-02", "sourceUrl": "https://www.anthropic.com/news/enterprise-frontier-safeguards" }, { "label": "software-03", "sourceUrl": "https://github.blog/changelog/2026-09-03-gemini-3-8-flash-is-now-available-in-github-copilot/" } ], "counterEvidence": "W36 had no production-volume next-generation silicon release: MediaTek-NVLink is a collaboration, Tensordyne is modeled pre-silicon, and Cerebras's 165 MW is a phased campus target with only the first phase under construction. The acceleration is clearest in software distribution and control-layer iteration, not uniformly across the hardware lens; two quarters of stalled production would still trigger the framework's refutation test." }, { "hypothesisNumber": 2, "hypothesis": "Capital is concentrated, returns are diffuse.", "verdict": "strained", "reasoning": "Large infrastructure commitments show capital concentration, including Nscale's delayed-draw facilities and the atNorth, UtilityInnovation Group, and EPC Power transactions. Salesforce packaging, Wonderful's valuation, and Atira's reported deployments do not measure diffuse returns or margins. The week supports the concentration half of the hypothesis but leaves the return-distribution half untested.", "evidence": [ { "label": "capital-01", "sourceUrl": "https://www.nscale.com/press-releases/nscale-3-billion-delayed-draw-term-loans" }, { "label": "capital-05", "sourceUrl": "https://www.vertiv.com/en-us/about/news-and-events/corporate-news/2026/vertiv-announces-agreement-to-acquire-utilityinnovation-group-to-accelerate-time-to-power-for-ai-data-centers/" }, { "label": "applications-03", "sourceUrl": "https://techcrunch.com/2026/09/02/wonderful-more-than-doubles-its-valuation-to-5b-in-under-6-months/" } ], "counterEvidence": "Broadcom captured $16.7 billion of quarterly AI-semiconductor revenue, 56.4% of company revenue, showing returns can concentrate with an upstream custom-silicon and networking supplier rather than diffuse downstream. Wonderful and Atira disclose funding, customer counts, and selected outcomes but not audited margins or retention, so application-layer return diffusion remains more plausible than measured." }, { "hypothesisNumber": 3, "hypothesis": "Networking is the durable layer.", "verdict": "strained", "reasoning": "W36 strengthened networking's strategic role: AWS-Azure private multicloud interconnect entered preview, DE-CIX expanded dual-homed ExpressRoute Metro, Equinix announced intent-driven Fabric One, and Firmus committed approximately $300 million to 25 years of APX East capacity. These events show orchestration, route diversity, and long-duration capacity value. They do not satisfy the framework's economic test because none discloses cross-connect or fabric revenue growth relative to compute revenue.", "evidence": [ { "label": "networking-04", "sourceUrl": "https://newsroom.equinix.com/2026-09-02-Equinix-Unveils-Equinix-Fabric-One,-Redefining-How-Enterprises-Connect-Across-AI,-Cloud-and-Networking-Infrastructure" }, { "label": "networking-05", "sourceUrl": "https://firmus.co/newsroom/firmus-and-subco-announce-major-trans-pacific-ai-connectivity-agreement" }, { "label": "networking-01", "sourceUrl": "https://aws.amazon.com/about-aws/whats-new/2026/08/aws-announces-AWS-interconnect-multicloud-microsoft-azure-preview/" } ], "counterEvidence": "Fabric One is pre-beta, the AWS-Azure preview currently reports up to 1 Gbps, and APX East targets Q4 2028. Meanwhile Broadcom's AI-semiconductor line combines networking with custom accelerators, preventing a clean margin comparison. The architectural evidence is favorable, but the week's public data cannot show networking pricing power holding longer than compute pricing power." }, { "hypothesisNumber": 4, "hypothesis": "Open weights pull the floor up.", "verdict": "untested", "reasoning": "NVIDIA's Apache-2.0 low-precision Muse-Glimmer checkpoint reduces deployment footprint while preserving published capabilities. But it is an optimization of an existing W33 model, not a new open-weight frontier release, and W36's dominant application and agent launches were closed or metered. One optimized checkpoint cannot test whether open weights narrowed the closed-model capability gap or rerouted meaningful new enterprise compute demand this week.", "evidence": [ { "label": "software-04", "sourceUrl": "https://huggingface.co/nvidia/Muse-Glimmer-30B-NVFP4" }, { "label": "applications-01", "sourceUrl": "https://www.salesforce.com/news/stories/salesforce-simplifies-editions-2026/" } ], "counterEvidence": "Astra's closed launch produced the week's strongest benchmark movement, and Salesforce, GitHub, and Anthropic all reinforced governed hosted access. The open floor may still be rising over a multiweek horizon, especially after W35's model wave, but W36 adds no comparable fresh benchmark set and should not be scored as support from absence of refutation." }, { "hypothesisNumber": 5, "hypothesis": "Power is the binding constraint for the next 24 months.", "verdict": "strained", "reasoning": "Capital and policy converged on energization: Vertiv and Flex agreed to power-control and conversion acquisitions; Google contracted 396 MW of geothermal with delivery beginning in 2028; Texas staff retained $50,000 per MW security for large-load interconnection; PJM's comment window closed on a bring-capacity-or-curtail proposal; and the Department of Energy temporarily authorized customer standby generation during emergency conditions. This supports power as a major site-timing constraint, but does not establish it as the unique or dominant constraint across the industry.", "evidence": [ { "label": "capital-05", "sourceUrl": "https://www.vertiv.com/en-us/about/news-and-events/corporate-news/2026/vertiv-announces-agreement-to-acquire-utilityinnovation-group-to-accelerate-time-to-power-for-ai-data-centers/" }, { "label": "capital-02", "sourceUrl": "https://www.stocktitan.net/sec-filings/FRVO/8-k-fervo-energy-co-reports-material-event-1de9aacddd08.html" }, { "label": "policy-02", "sourceUrl": "https://www.federalregister.gov/documents/2026/08/18/2026-16808/combined-notice-of-filings-2" }, { "label": "policy-03", "sourceUrl": "https://www.energy.gov/documents/doe-order-no-202-26-43" } ], "counterEvidence": "The Nscale facilities fund GPUs, networking, storage, cooling, and site work, not power alone, while Cerebras's Finnish build and the HUMAIN production cluster show projects can still secure viable sites. The evidence supports power as a site-timing constraint, not the only industry constraint; accelerator supply, financing, permits, and customer readiness still determine how fast announced capacity becomes revenue." } ], "patternWatch": [ { "pattern": "Agent deployment is shifting from capability access to explicit containment, delegated authority, and evidence custody.", "weeksObserved": 3, "instances": [ "W34: Anthropic's versioned-skills interface, Salesforce's externally callable data and agent services, and UiPath Maestro converged on identity-inherited agent orchestration.", "W35: The METR/OpenAI incident postmortems made evaluation-network isolation, scorer integrity, and default-deny tool boundaries central deployment requirements.", "W36: OpenAI attached action monitoring to Astra, Anthropic placed monitoring logs under customer keys, and GitHub made AI approval a scoped and revocable branch-protection state." ], "expectation": "Before 2026-12-31, at least one major enterprise agent platform will publish an exportable authorization and evidence schema covering tool scope, approvals, retained traces, and revocation. Falsified if the next two major agent releases provide capability GA without new authority or audit controls.", "reasoningType": "inductive" }, { "pattern": "Power constraints are migrating from site-selection assumptions into enforceable contracts, financial security, curtailment priority, and emergency dispatch.", "weeksObserved": 4, "instances": [ "W30: Georgia Power's 3.2 GW Project Camellia contract paired service with up to 1 GW of curtailment during grid stress.", "W34: Pennsylvania Executive Order 2026-05 applied new grid-readiness requirements above 25 MW, and a long-duration data-center transaction tied value to power availability.", "W35: Georgia formalized ratepayer safeguards, PJM proposed curtailment-first treatment for large loads without qualifying capacity, and Virginia began data-center electricity-tax collection.", "W36: PJM's comment deadline passed, Texas staff retained $50,000 per MW security above 75 MW, and DOE temporarily authorized customer standby assets as last-resort grid resources." ], "expectation": "By 2026-10-31, either FERC will act on PJM ER26-3515 or another US state or grid operator will publish a large-load rule that assigns measurable security, curtailment, or capacity obligations. Falsified if PJM's proposal is withdrawn and no comparable rule advances during that period.", "reasoningType": "inductive" }, { "pattern": "AI infrastructure buyers are reserving future schedule position through guarantees, capacity contracts, and milestone-timed capital rather than waiting for operating supply.", "weeksObserved": 3, "instances": [ "W34: A 20-year land-and-power shell transaction used an NVIDIA residual-value guaranty to make future accelerator-linked infrastructure financeable.", "W35: AWS reserved 2 million additional NVIDIA GPUs for 2027-2028, buying supply-chain position rather than reporting installed capacity.", "W36: Google contracted four 99 MW Fervo tranches from Q3 2028, Firmus committed to 25-year APX East capacity targeted for Q4 2028, and Nscale matched delayed-draw facilities to future deployment milestones." ], "expectation": "Before year-end 2026, another AI infrastructure contract above $500 million will disclose a future service date plus an option, earnout, guarantee, or delayed-draw mechanism that allocates schedule risk. Falsified if the next three comparable commitments are fully funded against currently operating capacity without milestone conditions.", "reasoningType": "inductive" } ], "secondOrder": [ { "trigger": "ARC Prize's matched maximum-reasoning comparison measured Astra at 62.7% on the Standard harness and 98.6% on the Provider Adapter, a 35.9-point spread associated with runtime state and context compression.", "effect": "Enterprise evaluations will have to version the model, harness, memory policy, tool surface, and retry budget as one tested system. Model-only scorecards will become procurement-incomplete, and portability reviews will ask which gains disappear when provider-private state is removed. ARC Prize, not this publication, established the core harness finding.", "horizon": "Q4 2026", "affected": "Enterprise AI architecture teams, model-evaluation vendors, frontier API providers, procurement and model-risk functions" }, { "trigger": "GitHub allowed Copilot approvals to count toward protected-branch requirements while Anthropic placed long-window monitoring evidence in customer-controlled accounts.", "effect": "Security and audit teams will define a new separation-of-duties rule: an AI may execute or approve, but the same provider-controlled evidence path cannot be the sole basis for both. Expect independent trace retention, human escalation thresholds, and policy-engine attestations in regulated software delivery.", "horizon": "Two procurement cycles", "affected": "Software engineering leaders, internal audit, regulated-industry security teams, GitHub administrators, agent-platform vendors" }, { "trigger": "Vertiv and Flex committed up to $7.0 billion to grid-to-chip controls and power conversion while large-load rules attached security and curtailment obligations to interconnection.", "effect": "Data-center design authority will move earlier toward firms that can model utility, onsite generation, storage, power conversion, and rack loads as one permitted system. Cooling and compute vendors without an upstream power-control partner will face acquisition, partnership, or specification risk before equipment selection.", "horizon": "2027 planning cycle", "affected": "Data-center developers, electrical equipment vendors, utilities, engineering firms, hyperscalers, neoclouds, and infrastructure investors" } ], "strategicOutlook": "W36 favors control-plane and schedule discipline over headline capacity. Treat Astra as a model-plus-runtime release: reproduce its benchmark advantage under the memory, context-compression, tool, and audit constraints you can govern. For agent procurement, forecast entitlements and credits against completed workflow outcomes rather than seats or tokens, and require exportable authorization and evidence records before allowing AI approvals to satisfy human control gates. For infrastructure, separate operating megawatts from phased, optioned, and contracted future capacity; track dated service paths, then assign delivery risk through earnouts, delayed draws, guarantees, or tranche milestones. Power is a major site-timing constraint, not a proven unique bottleneck. Networking's role is strengthening architecturally, yet the thesis needs revenue evidence before claiming durable pricing power." }, "predictions": [ { "id": "p100-astra-neutral-memory-dec15", "text": "ARC Prize or another provider-neutral evaluator publishes an Astra run without provider-private reasoning state that closes at least half of the 35.9-point matched-effort Standard-to-Provider-Adapter gap by December 15, 2026.", "confidencePct": 32, "deadline": "By December 15, 2026", "trigger": "Public ARC-AGI-3 result with a reproducible visible-memory and context-compression configuration, no provider-private reasoning state, and Astra at 80.7% or higher.", "lens": "software" }, { "id": "p101-fervo-expansion-firm-jun30", "text": "Google accepts at least 500 MW of Fervo's conditional expansion and the parties execute a definitive agreement by June 30, 2027.", "confidencePct": 41, "deadline": "By June 30, 2027", "trigger": "Fervo filing or Google announcement stating that at least 500 MW of the approximately 600 MW expansion option has become firm under a definitive agreement.", "lens": "power" }, { "id": "p102-second-queue-contract-dec31", "text": "A second AI infrastructure contract above $500M discloses both a future service date and an option, delayed-draw, earnout, or guarantee allocating schedule risk by December 31, 2026.", "confidencePct": 68, "deadline": "By December 31, 2026", "trigger": "Primary filing or announcement with transaction value above $500M, named service date, and explicit contingent or milestone mechanism.", "lens": "capital" }, { "id": "p103-fabric-one-control-schema-dec31", "text": "Equinix publishes Fabric One beta documentation that exposes approval, rollback, or auditable intent-history controls before December 31, 2026.", "confidencePct": 57, "deadline": "By December 31, 2026", "trigger": "Public Fabric One beta documentation naming at least one of policy simulation, approval workflow, rollback, or exportable intent history.", "lens": "networking" }, { "id": "p104-broadcom-ai-split-dec31", "text": "Broadcom discloses separate quarterly revenue figures for custom AI accelerators and AI networking by December 31, 2026.", "confidencePct": 24, "deadline": "By December 31, 2026", "trigger": "Hit only if a Broadcom earnings release, 10-Q, or call transcript reports distinct dollar revenue for both custom AI accelerators and AI networking; a combined AI-semiconductor line is a miss.", "lens": "hardware" } ], "predictionsPrior": [ { "id": "p95-jalapeno-agentx-oct31", "text": "SemiAnalysis publishes AgentX v3 multi-turn benchmark results for OpenAI Jalapeño on production-representative agentic traces, with methodology comparable to Vera Rubin NVL72 AgentX runs cited by NVIDIA, by October 31, 2026.", "confidencePct": 36, "deadline": "By October 31, 2026", "trigger": "SemiAnalysis newsletter or InferenceX page listing Jalapeño AgentX throughput-per-megawatt and cost-per-million-tokens on multi-turn traces, not solely 8k/1k InferenceX STP runs.", "outcome": "pending", "notes": "Resolution window remains open; no comparable AgentX publication in the validated W36 evidence.", "lens": "hardware" }, { "id": "p96-openrouter-volume-sep30", "text": "The highest single ISO week of OpenRouter aggregate token volume in September 2026 exceeds the Ox Alpha stealth-week peak (week of August 20–26, 2026) by at least 15%, by September 30, 2026.", "confidencePct": 40, "deadline": "By September 30, 2026", "trigger": "OpenRouter public stats page or Requesty/OpenRouter blog post reporting weekly tokens processed for each September 2026 ISO week against the Ox Alpha peak week — week versus week, same unit. The baseline peak week ran at zero list price and is community-reported (grade 3 on our 1–5 source scale), so the comparison inherits that grade.", "outcome": "pending", "notes": "September is incomplete at the September 5 cutoff; a highest-week comparison cannot yet be scored.", "lens": "software" }, { "id": "p97-bis-remote-gpu-nprm-nov30", "text": "Commerce BIS publishes a Federal Register notice of proposed rulemaking on remote access to advanced US AI compute by Chinese end users, by November 30, 2026.", "confidencePct": 27, "deadline": "By November 30, 2026", "trigger": "Federal Register NPRM from Commerce/BIS with docket number and comment period addressing remote GPU access via third-country data centers.", "outcome": "pending", "notes": "Resolution window remains open; no qualifying Federal Register notice appears in the validated W36 evidence.", "lens": "power" }, { "id": "p98-nvidia-rubin-mix-q3-earnings", "text": "NVIDIA Q3 FY2027 earnings disclosure states Vera Rubin contributed more than 25% of datacenter revenue for the quarter ended October 26, 2026.", "confidencePct": 74, "deadline": "By NVIDIA Q3 FY2027 earnings release (expected November 2026)", "trigger": "NVIDIA Form 10-Q or earnings call transcript for quarter ended October 26, 2026 stating Vera Rubin datacenter revenue mix above 25%.", "outcome": "pending", "notes": "The predicted quarter has not ended and the earnings disclosure has not occurred.", "lens": "hardware" }, { "id": "p99-metaroce-ocp-spec-oct31", "text": "Meta contributes MetaRoCE specification through OCP at the October 2026 Global Summit with documented production deployment targets beyond the 64-node AMD proof-of-concept, by October 31, 2026.", "confidencePct": 44, "deadline": "By October 31, 2026", "trigger": "OCP Global Summit 2026 materials or Meta Engineering blog publishing MetaRoCE spec with named hyperscaler or cloud deployment timeline distinct from the August 64-node lab cluster.", "outcome": "pending", "notes": "The October summit and deadline remain ahead; no qualifying production target appears in W36 evidence.", "lens": "networking" } ], "watchlist": [ { "window": "Sep 8", "title": "DOE emergency-order expiry", "why": "Order 202-26-43 expires after temporarily authorizing customer standby generation as a last-resort grid resource." }, { "window": "Sep 9", "title": "GLM-5.3-Flash promotion ends", "why": "Reprice W35 open-weight agent pilots on steady-state list rates rather than the launch discount." }, { "window": "Sep 11", "title": "Texas large-load rule vote", "why": "Tests whether the 75 MW threshold and $50,000/MW security move from staff recommendation into final policy." }, { "window": "Later in 2026", "title": "Fabric One beta documentation", "why": "Look for policy simulation, approval, rollback, service levels, and exportable intent evidence before production use." }, { "window": "Q4 2026", "title": "Vertiv-UIG and Flex-EPC closings", "why": "Regulatory clearance and disclosed integration terms will test whether grid-to-chip consolidation converts from agreement to operating capability." } ], "changelog": [ "W36 authored from validated research, counterbrief, house measurement, and synthesis artifacts through the September 5 cutoff.", "Three model-tree rows added: GPT-6 Astra, Claude Fable 5.1, and Gemini 3.8 Flash.", "All five W35 predictions remain pending because their resolution windows have not matured." ] } }