{"product":"ai-compute-strategic-delta-feed","surface":"track-record","asOf":"2026-07-27","windowDays":365,"delayDays":7,"scorecard":{"n":0,"overallHitRate":null,"overallBrier":null,"overallBaseRatePrior":null,"overallModelVsBaseline":null,"thin":true,"byGrammar":[]},"resolvedCalls":[{"insightId":"804cbf50-8304-45d8-babf-8f08cc4c76a5","headline":"Federal Register AI-compute publications accelerated 178% this week, contradicting the consensus view that regulatory pace won't materially shift near-term assumptions.","grammar":"narrative_lag","niche":"strategic_major","conviction":0.85,"verdict":"inconclusive","resolvedAt":"2026-07-19T23:55:46.100Z","trippedSignpost":null,"birthSealHash":"122cd4768985830342d212e75dae7633f050d5e169aa9d2c0e68eb6a67774bd9","citation":"luu:insight:804cbf50-8304-45d8-babf-8f08cc4c76a5#122cd4768985","birthSealedAt":"2026-07-02T08:35:28.398Z","birthSealed":true,"resolutionSealHash":"bf648f60f9ad44f48e736ffc7b782dec005de851cbc271f2ae1a47f392deaa42","resolutionSealedAt":"2026-07-19T23:55:46.100Z","resolutionSealed":true,"forwardTested":false,"evidenceLeadDays":null,"entities":[]},{"insightId":"8707fea9-72de-4a8a-b89e-9c307d467fe2","headline":"Federal AI-compute regulatory publishing accelerated 227% this week, challenging the consensus view of slow-moving regulators.","grammar":"narrative_lag","niche":"strategic_major","conviction":0.85,"verdict":"confirmed","resolvedAt":"2026-07-19T23:55:45.363Z","trippedSignpost":{"description":"Over the next 14 days, AI-compute-relevant regulatory publication maintains the ≥50% acceleration (≥29 new docs per 7-day window)","expectedDirection":"supports","outcome":"tripped_supports","trippedAt":"2026-07-02T00:00:00.000Z"},"birthSealHash":"4a97df9a019e5a19ca59a5f833abbaf0457d4eeda7eff533b638bc6a9ee87a54","citation":"luu:insight:8707fea9-72de-4a8a-b89e-9c307d467fe2#4a97df9a019e","birthSealedAt":"2026-07-03T21:55:12.500Z","birthSealed":true,"resolutionSealHash":"40b8a5786469af1871136e3be60ccbfc1be4713521b113c963d19a44dfa3e56b","resolutionSealedAt":"2026-07-19T23:55:45.363Z","resolutionSealed":true,"forwardTested":false,"evidenceLeadDays":-1.45,"entities":[]},{"insightId":"399a96e3-5ceb-47d7-a862-7af5eb12faf8","headline":"The assumption that GPU allocation remains the primary bottleneck for frontier training faces mounting contradictory signals. Three portfolio decisions depend on this increasingly uncertain premise.","grammar":"quiet_assumption_invalidation","niche":"strategic_major","conviction":0.769,"verdict":"falsified","resolvedAt":"2026-07-19T23:55:42.269Z","trippedSignpost":{"description":"Proxy eia.miso.demand_hourly_mw reverts within 1σ 21d","expectedDirection":"falsifies","outcome":"tripped_falsifies","trippedAt":"2026-06-24T12:00:00.000Z"},"birthSealHash":"ff29bd5cd7e2c0b726f7c661a9dd1f5a8adfcc619699cefea39e9b48ed16bbf0","citation":"luu:insight:399a96e3-5ceb-47d7-a862-7af5eb12faf8#ff29bd5cd7e2","birthSealedAt":"2026-07-02T08:35:27.888Z","birthSealed":true,"resolutionSealHash":"e9fcda419104f05efd1a271695eb752c8eb52cbb3572b20061eaa4a90a97bbed","resolutionSealedAt":"2026-07-19T23:55:42.269Z","resolutionSealed":true,"forwardTested":false,"evidenceLeadDays":-7.82,"entities":[]},{"insightId":"fa919bde-34b5-471a-ab7e-e5695404ba3d","headline":"The assumption that GPU allocation remains the primary bottleneck for frontier training faces mounting pressure from two contradictions, with three portfolio decisions at stake.","grammar":"quiet_assumption_invalidation","niche":"strategic_major","conviction":0.582,"verdict":"falsified","resolvedAt":"2026-07-19T23:55:41.234Z","trippedSignpost":{"description":"Proxy eia.miso.demand_hourly_mw reverts to within 1σ of its 90-day baseline within 21 days — the pressure on the assumption relaxes.","expectedDirection":"falsifies","outcome":"tripped_falsifies","trippedAt":"2026-06-24T12:00:00.000Z"},"birthSealHash":"844629b839558b3a91424d5a037c1e08529421dabce7a52f741bd3537d4f63aa","citation":"luu:insight:fa919bde-34b5-471a-ab7e-e5695404ba3d#844629b83955","birthSealedAt":"2026-07-03T21:55:12.378Z","birthSealed":true,"resolutionSealHash":"866adc7c3d8ca767662c7c474b1f498331f1cbbbff8e0afd1d0daa47ddd9362c","resolutionSealedAt":"2026-07-19T23:55:41.234Z","resolutionSealed":true,"forwardTested":false,"evidenceLeadDays":-8.95,"entities":[]},{"insightId":"508316d3-be79-4a8f-8c76-bc36b175895e","headline":"NVIDIA TensorRT-LLM commit activity drops 4σ below baseline, pressuring assumptions about typical development velocity.","grammar":"quiet_assumption_invalidation","niche":"strategic_major","conviction":0.36,"verdict":"falsified","resolvedAt":"2026-07-02T11:03:17.149Z","trippedSignpost":{"description":"Within 14 days, github.NVIDIA.TensorRT-LLM.commits_7d reverts to within 1σ of baseline (mean ≈ 97.89)","expectedDirection":"falsifies","outcome":"tripped_falsifies","trippedAt":"2026-06-19T08:31:17.451Z"},"birthSealHash":"080811817b9722254196f769678172003806309d2a852672514f5c5ce381d164","citation":"luu:insight:508316d3-be79-4a8f-8c76-bc36b175895e#080811817b97","birthSealedAt":"2026-07-02T08:35:28.188Z","birthSealed":true,"resolutionSealHash":"386116e9ce4bd8a97163f70b05a8656acd432fcad4d1a8e722d81682bf474480","resolutionSealedAt":"2026-07-25T16:18:29.353Z","resolutionSealed":true,"forwardTested":false,"evidenceLeadDays":-12.96,"entities":[]}],"forwardTestedCount":0,"retrodictiveCount":5,"withdrawnCriterionCount":78,"withdrawnCallCount":46,"withdrawalNotice":"On 2026-07-25 we withdrew 78 machine-checkable resolution criteria across 46 open calls, before any of them resolved. Most were sigma-anomaly tests over hourly US grid demand — a comparison that cannot separate the claim being tested from summer air-conditioning load and the daily demand cycle. Two independent errors: the criteria had no stable operating characteristic (one pair's confirming side would have fired on 0.44% of scans measured on 2026-07-10 and 11.65% on 2026-07-25, with the falsifying side moving the opposite way, on an unchanged thesis), and a \"3 sigma\" criterion scanned hourly for 30 days offers 526 chances to fire, so it has roughly a 51% chance of firing on noise alone against the 0.135% its label implies. The affected calls remain open and visible; they are not scored, and they stay in the denominator. The withdrawal rule is arithmetic that was available on the day each criterion was sealed, was applied to every criterion that failed it regardless of which way it was trending, and was recorded while the resolution monitors were disabled — so the sealed record shows the withdrawals preceded any outcome.","methodology":{"preregistered":true,"sealedAtBirth":true,"sealedAtResolution":true,"forwardTestRule":"A call is SCORED only when the evidence that resolved it postdates the sealed claim. Resolutions whose evidence predates their own seal are retrodictive — the outcome was already determined before the belief was committed — so they are published (with a signed evidenceLeadDays) but excluded from the scorecard, exactly as a registered report excludes outcomes measured before the protocol was locked.","scoring":"Brier score + hit-rate with sample-size-aware (beta-binomial) 90% confidence intervals. confirmed/partial = hit; falsified = miss; aged_out/inconclusive carry no outcome.","criterionAdmissibility":"A resolution criterion must be able to discriminate the claim it is attached to. We withdraw criteria whose realized firing rate on the metric's own history — measured with no reference to the claim — shows they are driven by something else: a strong seasonal or diurnal cycle, a censored series where the threshold is unreachable, or too few distinct values for a sigma to mean rarity. We also correct for the fact that a criterion scanned repeatedly gets many chances to fire, so a nominal '3 sigma' test evaluated hourly for a month is not a 3-sigma test. Withdrawn criteria are preserved unmodified, sealed with the reason, and their calls stay open and visible in the denominator.","verifyAnInsight":"GET /api/public/insight-integrity-proof?insightId={id} returns a Merkle inclusion proof for each sealed lifecycle record, plus the anchored root's public commit — no API key required. (The richer evidence pack at POST /api/v1/insights/{id}/evidence-pack needs an API key.)","publicDelay":"Public tier reflects data resolved ≥ 7 days ago. n counts FORWARD-TESTED calls only; retrodictive calls remain visible in resolvedCalls but are not scored, so n equals forwardTestedCount, not the total number of calls shown.","protocolDoc":"/proof/scbsco-reference-run-001","anchors":"https://github.com/cberzins1973/luu-attestation-anchors"},"formingNotice":"5 call(s) have resolved, but none yet meets our scoring standard (forward-tested AND soundly adjudicated): the evidence that resolved them predated the sealed claim, so we score none of them. Each is listed below with its signed evidenceLeadDays. The scored record begins with the first resolution whose evidence postdates its seal.","notice":"Research demonstration corpus. SCB/SCO Reference Run #001 — AI Compute (30 days). Not investment advice, not a commercial product, not a claim of complete coverage. See /proof/scbsco-reference-run-001 for the pre-registered protocol and methodology.","traceId":"phFXfpcttAYGVzNqJ1l2D"}