as of 2026-05-08

Related Work

Peer organisations active in compute-governance research and adjacent domains: Epoch on facility-level frontier verification, CSET on export-control and chip supply-chain analysis, RAND on firm-network systemic risk, Stanford HAI on AI capability and investment indices, EPRI on data-centre power demand, GovAI on training-compute thresholds and frontier policy. The sections below name, for each, the specific artefact (a join, a column, an edge type, or a refresh cadence) this dataset adds.

Peer organisations active in compute-governance research and adjacent domains: Epoch on frontier facilities, CSET on export controls and chip supply chains, RAND on systemic risk, Stanford HAI on AI capability and investment indices, EPRI on data-centre power, GovAI on compute thresholds. The sections below name what Scrutica adds for each.

Peer organizations

Peer research network

Seven organizations whose work Scrutica cites and extends. Epoch AI is the closest peer on AI compute infrastructure; the rest occupy adjacent territory (export controls, firm-network systemic risk, AI capability + investment indices, frontier-capability evaluation, power-demand research, governance scholarship). Each entry links to the flagship dataset or paper Scrutica references most often.

The peer organizations whose work Scrutica builds on. Epoch AI publishes the most rigorous frontier-facility dataset in the field; the others cover adjacent layers (export controls, supply-chain risk, the AI Index, frontier-capability evaluation, grid impact, governance research). Each entry names the flagship publication Scrutica cites.

Epoch AI

Frontier compute measurement; the closest peer for AI compute infrastructure.

Epoch's Frontier Data Centers methodology, anchored on satellite imagery (SkyFi, Sentinel-2, Google Earth), permitting documents, and corporate disclosures, is the verification benchmark Scrutica reaches for whenever per-facility depth is the discussion. Epoch's own published uncertainty framing on that methodology page is bounded to cooling-model error (up to 2× theoretical; 1% on the two ground-truth cases they hold; ≤50% against indirect references such as chip-quantity-derived IT power) and does not quantify confidence bounds on compute, cost, or timing. Epoch's AI Chip Owners explorer, ML Hardware database, and Notable AI Models corpus feed several Scrutica computations directly (chip-vendor cross-validation, FLOP/s + TDP constants, the model side of the threshold-atlas).

Epoch AI publishes the most rigorous facility-level dataset in the field: 13 frontier US data centers verified against satellite imagery and permits. Their stated uncertainty is bounded to cooling-model error (up to 2× theoretical, ≤50% in practice against indirect references); they do not publish quantified confidence bounds on compute, cost, or timing. Scrutica ingests and attributes their work, then adds substrate they do not currently cover (sovereign programs, supply-chain edges, export-control designations).

Frontier Data Centers Hub

CC-BY 4.0; 13 facilities; ~15% of global delivered AI compute as of November 2025.

CSET (Georgetown)

Export-control + chip supply-chain analysis.

The Center for Security and Emerging Technology's Emerging Technology Observatory ships the Advanced Semiconductor Supply Chain Dataset (May 2025), drawing on TechInsights CMRS, WSTS, and SIA, with per-input market shares and headquarters of ultimate parents. The dataset is a static viewer, not a propagation engine; Scrutica's cascade and concentration explorers operate on a broader edge corpus (licensed supply-chain databases + CSET ETO + a licensed financial-fundamentals database + chip-deployment chains, 20,534 edges total).

CSET's policy briefs and Chip Explorer dataset frame the export-control analysis Scrutica's BIS cross-reference and coordination-gap features build on. Scrutica's contribution is the cascade engine that propagates a designation through the supply graph and the gap analyzer that maps which allies can enforce coordinated controls.

CSET ETO Advanced Semiconductor Supply Chain Dataset

RAND Corporation

Firm-network systemic-risk research.

RAND working paper WR-A2625-1 (Welburn et al., 2023) is the closest published methodology peer for Scrutica's cascade simulator: the same licensed bilateral-relationship database underpins both, with ML-imputed missing edges. RAND's published RAND-on-semiconductors corpus (PE-A1394-1, RR-A2354-1, RR-4185, WR-A2625-1) does not publish a per-hop propagation-decay parameter directly comparable to Scrutica's; the validation cross-walk frames Scrutica's heterogeneous-criticality cascade as distinct from, rather than a replacement of, RAND's input-output linkage estimation.

RAND's Taiwan-strait scenario analyses and microelectronics policy work are the institutional reference for stress-testing the supply chain. Scrutica's Cascade Simulation operationalizes this as an interactive engine: a user selects a chokepoint and the simulation walks the failure forward through the supply graph, surfacing which downstream nodes survive on inventory buffers and which substitute under what timeline.

The Global Economy at the Firm-Level (Welburn et al., 2023)

Stanford HAI

AI Index: the institutional reference for AI capability + investment metrics.

The Stanford Institute for Human-Centered AI's AI Index aggregates training-compute, capex, and capability-benchmark trends across the field, with explicit acknowledgment of Epoch AI as a primary data provider for the compute and capability sub-indices. Scrutica's coverage of $454.4M in AI-related public contracts cited on the Sovereign AI dashboard's HAI peer column traces to AI Index 2026 footnote 10 directly.

Stanford HAI's AI Index is the citation standard policymakers reach for when they need a stocktake on training compute, AI capabilities, and AI investment. Scrutica's Sovereign AI Reality Dashboard cites HAI's per-country investment numbers and reconciles them against announced-vs-deployed program execution, a layer HAI does not produce directly.

AI Index Report 2026

METR (Model Evaluation & Threat Research)

Frontier-capability evaluation; the empirical bound on what compute thresholds mean.

METR's HCAST methodology ("Measuring AI Ability to Complete Long Tasks", Kwa et al., 19 March 2025) and the Time Horizons 1.1 dataset (released 29 January 2026) are the primary surface Scrutica's Autonomy Capability Threshold Monitor reads from: the autonomy-monitor module sets METR_DATA_HUB_URL to metr.org/time-horizons/, carries v1.1 entries through 2026-04-07, and last refreshed against the YAML on 2026-05-09. METR does not produce training-FLOP estimates; the FLOP enrichment on Scrutica's autonomy-monitor surface joins per-model to Epoch AI's training-compute database, asking the question neither org produces in isolation — what training compute gates which 50%-time-horizon. The relationship is upstream-citation rather than extension: Scrutica reads METR's published time-horizons at face value (authority_tier 2) and adds the per-facility complement (which facilities have the training capacity to enter which time-horizon band).

METR is the independently funded nonprofit (no AI-company funding; supported by The Audacious Project, the Sijbrandij Foundation, and individual donors) whose autonomous-task evaluations bound what compute thresholds mean policy-wise. A 10²⁵ FLOP line (EU AI Act) or 10²⁶ FLOP line (rescinded US EO 14110; NY RAISE Act 10²⁶ + $500M revenue, effective 2027-01-01) is only policy-meaningful when paired with empirical capability bounds; METR's time-horizon work (the frontier 50%-horizon doubling roughly every 7 months from 2019 onward) is the citation Scrutica reaches for whenever the Threshold Atlas needs the capability complement to the compute reading. METR partners with the UK AI Security Institute, the US AISI Consortium, OpenAI, and Anthropic for pre-release model evaluations.

Measuring AI Ability to Complete Long Tasks (Kwa et al., March 2025)

HCAST + RE-Bench + SWAA suite; v1.1 dataset refreshed January 2026; metr.org/time-horizons/ is the live YAML Scrutica's autonomy-monitor refreshes against.

EPRI

Power-demand and grid-impact research for data centers.

The Electric Power Research Institute publishes facility-level power-demand methodology used by US utility planners and grid operators, including PUE conventions and load-curve work that informs Scrutica's Cost Index decomposition (chip TGP × interconnect × MFU × hours-utilized). EPRI's 2024 data-center-power study is the canonical reference Scrutica's PUE anchor table cross-checks against (default 1.15; range 1.05–1.45).

EPRI is the bridge between AI compute infrastructure and the electric grid. Their data-center load research informs Scrutica's Compute Cost Index and the energy-side framing on facility detail pages: what the grid has to deliver, where queues bind, where heat-recovery and behind-the-meter generation reshape the regional picture.

Powering Intelligence: EPRI Data Center Energy Research

GovAI (Centre for the Governance of AI)

Governance research: compute thresholds, frontier policy, allied coordination.

GovAI's research on compute governance (training-compute thresholds, on-chip mechanisms, verification regimes, frontier-AI scaling curves) is the academic counterpart to Scrutica's threshold-atlas and coordination-gap surfaces. Scrutica operationalizes the empirical question (which facilities can train above a 10²⁵ FLOP cumulative threshold, under whose jurisdiction, with what enforcement reach) that GovAI's methodological work frames.

GovAI is the academic centre studying the governance of advanced AI. Their work on compute thresholds (the EU AI Act 10²⁵ line, the China CAC line, on-chip verification) shapes the policy register Scrutica's Threshold Atlas and Coordination Gap analyzer translate into facility-level findings.

GovAI Research Hub

Scrutica-only substrate

Where Scrutica extends

The substrate Scrutica produces and the engines built on top of it. Each row links to the live feature page; the methodology behind each lives at /methodology with the full derivation chain.

The features Scrutica ships that peer organizations don't. Click any row to land on the feature; the full methodology behind each is at /methodology.

/supply-chain

Supply-chain edges (licensed supply-chain databases + CSET ETO + financial-fundamentals)

20,534 weighted directed edges with relationship-rank-derived criticality (rank 1–3 → bucket 8; 4–10 → 6; 11–25 → 4; 26+ → 3) and 3-month price-correlation fields populated for 78% of licensed-database rows. Drives /cascade propagation, /chokepoint-anatomy, /coordination-gaps reach analysis, and the regulatory-impact pre-scorer.

The supplier-customer relationships among the firms that build, operate, and sell AI compute. Drives the cascade simulation and the concentration analysis: what propagates when a chokepoint breaks, and where the supply chain narrows to single suppliers.

/sovereign-ai

Sovereign AI program tracking

34 programs with announced-vs-committed-vs-disbursed columns, per-country derivation chain (facility / chip allocation / utility), inflation decomposition (government-only vs. private + FDI), authority-tier flags on every cited figure, and a public-records traceability tab walking licensed-database LP→fund→portfolio chains where available.

Government AI programs across 34 countries, with the announced number, the committed number, and the disbursed number tracked separately. The page surfaces the gap between aspiration and execution honestly: France's 97% private-capital framing of the €109B Action Summit pledge, Saudi Arabia's PIF-financed opacity, South Korea's ₩100T (~$74B) headline vs. $1.481B government-only AI spending ($381M 5-consortia model program + $1.1B NAICC compute).

/export-controls

Export-control designations + cascade

3,425 Federal-Register-anchored designations + a closure walker that propagates structural exposure to compute-universe organizations via NAME_SIMILARITY matching and AFFILIATE_CLOSURE BFS through ownership_chain_edges. Honest about the BIS Affiliates Rule suspension (effective Nov 10, 2025 through Nov 9, 2026): cascade signal is heightened-due-diligence trigger, not implied designation.

Every Entity List, MEU List, and SDN designation tracked, with cross-references against the 4,538-facility / 24,783-organization compute-relevant substrate (121,851 canonical-resolved organizations in the universe). When BIS designates an entity, Scrutica's cascade walks the affiliate chain and surfaces which compute-relevant organizations are structurally exposed.

/cost-index

Compute Cost Index ($/petaFLOP-day)

Daily Vercel cron refreshes AWS/Azure/Oracle on-demand prices (vendor-canonical APIs); GCP/CoreWeave/Lambda refreshed via Playwright-verified manual ingest. Prices normalized to $/petaFLOP-day under the dense BF16 convention (no 2:4 structured sparsity; H100 SXM5 = 989.5 TFLOP/s) anchored to Epoch's ML Hardware database.

What it costs to train or serve at unit scale across the major cloud providers, broken down by hardware generation. The denominator is petaFLOP-days (one petaFLOP sustained for one day), so the comparison is normalized across H100, H200, B200, MI300X, GB200.

/cascade

Cascade Simulation (animated, weighted)

BFS forward + reverse + bidirectional propagation over the 20,534-edge graph. Per-scenario propagation decay (0.65–0.95) layered onto per-edge `flowShare × criticality/10` multiplier; six prebuilt scenarios (TSMC disruption, ASML EUV halt, US-China decouple, CoWoS bottleneck, HBM disruption, allied-only EUV) plus freeform mode. Validated against 2 of 3 historical events at qualitative top-5 reconciliation (Stargate Abilene RECONCILED; CoWoS 2023 RECONCILED with parameter-provenance transparency).

Pick a chokepoint (TSMC, ASML, SK Hynix, Nvidia) and watch the failure propagate through the supply graph. The simulation weights edges by licensed-database relationship rank and per-scenario decay; downstream nodes survive on inventory buffers (ASML EUV cliff at zero weeks; advanced wafers linear at 10 weeks; HBM cliff at 3 weeks) and recover via parameterized substitution paths.

/threshold-atlas

Threshold Monitor (EU AI Act, China CAC, EO 14110)

Facility-level FLOP capacity vs. each regulatory threshold (EU AI Act 10²⁵ cumulative; China CAC ~10²⁴; rescinded EO 14110 10²⁶) with provenance click-through on every numeric cell. Path-A hardware estimate when GPU inventory is known; Path-B power-derived estimate (facility MW / PUE × GPU-fraction-of-IT / chip TGP) when only nameplate is known.

Which AI data centers around the world have enough compute to put a model across each regulatory line, under whose jurisdiction, with what enforcement reach. Pairs the model-side question Epoch's Notable Models data answers (cumulative training compute per model) with the facility-side question Epoch does not.

/compute-visibility

Compute Visibility Index by country

Country-level deployment with hosted-vs-controlled attribution (capacity sitting on soil vs. capacity controlled by domestic-incorporated owners) and ownership-transparency tiering across the global footprint. Cross-validates against Epoch's company-level 9-entity chip-owners series (Google, Microsoft, Meta, AWS, China-aggregate, Oracle, CoreWeave, xAI, Other).

How much AI compute each country actually controls, distinct from how much sits on its territory. The first public estimate of 'compute under control' by jurisdiction; the rollup of Epoch's company-level series into country-level totals where ownership chains resolve cleanly, with explicit ownership-opacity flags where they don't.

/coordination-gaps

Allied Coordination Gap Analyzer

Per-allied-country reach over each chokepoint tier (lithography → wafer → packaging → memory → accelerator → cloud), with explicit flagging of where the allied legal authority for coordinated export controls breaks down (Korean MSIT semi-coordination via WA Multilateral but not fully aligned on Entity List; Singapore's narrower extraterritorial framework; Taiwan-specific gaps).

Which US allies have the legal authority to coordinate export controls on which chokepoints. Where the gaps fall (Singapore, Korea, Israel for some technologies), the page surfaces the structural reason and the policy lever that would close it.

/methodology

Methodology Explorer (per-cell click-through)

Every numeric cell rendered through the ProvenanceValue infrastructure carries the full envelope on click: data_source, source_url, vintage, authority tier (T1 primary measurement → T4 editorial assessment), method, confidence, and any documented conflicts. Sliders on the FLOP engine and cascade page propagate parameter changes through the entire derivation chain.

Click any number on Scrutica and read where it came from. Every value carries its source URL, when the data was originally produced, the authority tier (primary measurement, research database, press, estimate), and the conflicts where sources disagree, surfaced inline rather than buried in a methodology footnote.

Coverage limits

Outside Scrutica's scope

What Scrutica explicitly does not cover, with the peer organization that does. Knowing where the line falls is load-bearing for citing the data correctly. None of these are gaps queued for later coverage; they are scope decisions that hold for the foreseeable horizon.

What Scrutica explicitly does not do, and where to go instead. None of these are gaps queued for later coverage; the scope decisions hold for the foreseeable horizon.

Per-facility satellite imagery + permit primary investigation

Scrutica does not run independent satellite-imagery analysis or permitting-record investigation on individual facilities. Where we ingest Epoch's frontier-DC rows (data_source: epoch-ai), we read their work at face value rather than re-investigating. The verification benchmark for the 13 frontier facilities Epoch covers is theirs.

Epoch AI: Frontier Data Centers methodology (the citation standard)

AI Models corpus (cumulative training compute per model)

Scrutica does not catalog ML models. The model side of the threshold question (which trained system has crossed which compute line) is Epoch's Notable Models database; Scrutica's Threshold Atlas asks the facility-side complement and links into the Notable Models corpus where appropriate.

Epoch AI: Notable AI Models

AI Companies revenue / funding / staffing tracking

Scrutica does not track AI-company revenue, headcount, or rolling funding totals. The AI Companies dataset Epoch publishes (revenue, funding, staff, usage signals with 'Confident' / 'Likely' tier per data point) is the citation source for company-level commercial trajectory; Scrutica's company pages surface compute-relevant signals (deployments, ownership chains, supply-chain edges) only.

Epoch AI: AI Companies

AI capability benchmarking (METR HCAST, ARC-AGI, FrontierMath, etc.)

Scrutica does not run capability benchmarks or curate evaluation suites. METR's autonomous-task length curve, the FrontierMath benchmark, the ARC-AGI corpus: capability work belongs to the labs and benchmark organizations doing it. Scrutica's Capability Scaling explorer cites their findings rather than reproducing them.

METR (Model Evaluation and Threat Research): capability benchmarking

AI Index (the comprehensive annual stocktake)

Scrutica does not publish an annual AI Index. Stanford HAI's AI Index is the field's institutional reference for the broad question 'where is AI overall'; Scrutica is the narrower compute-infrastructure layer that AI Index pulls from for its training-compute and capex sections.

Stanford HAI: AI Index

Classified or proprietary procurement data

Scrutica relies on public-records substrate (USAspending, EU TED, simpler.grants.gov, Federal Register, SEC EDGAR, vendor disclosures) plus licensed corporate-ownership and supply-chain databases held under subscription. Classified contracts (intelligence community, advanced military procurement) sit outside the public record by design and outside Scrutica's coverage by construction. Where the public record carries gaps that distort the picture, the methodology pages flag them explicitly.

USAspending.gov: federal procurement public-records substrate

Real-time grid-load measurement at the data-center bus

Scrutica tracks announced power-capacity nameplate (in MW) and queued grid-interconnection requests where filed publicly (ERCOT, MISO, PJM queues; published utility data). We do not run live grid-load telemetry against individual facilities; that work is utility-internal and EPRI-adjacent.

EPRI: Data Center Energy Research

Cite this page

Scrutica. "Related Work: How Scrutica relates to Epoch AI and the broader compute-governance research network." 2026-05-08. https://scrutica.com/related-work

When citing this page, include the date so readers can map the comparison to a specific snapshot of substrate counts; the live counts appear in the "as of" stamp at the top.

METR (Model Evaluation & Threat Research)

Frontier-capability evaluation; the empirical bound on what compute thresholds mean.

Measuring AI Ability to Complete Long Tasks (Kwa et al., March 2025)

HCAST + RE-Bench + SWAA suite; v1.1 dataset refreshed January 2026; metr.org/time-horizons/ is the live YAML Scrutica's autonomy-monitor refreshes against.