GPU lead times extend to 18+ months; hyperscaler capex exceeds $400B/year; power grid investment lags demand; AI model costs stay high; enterprise adoption slows relative to expectations.
Signal counts measure media attention over the last 7 days — not the likelihood of an outcome.
If this branch plays out and you weren't positioned, here's what you'd miss or take. AI-generated estimates, not forecasts.
▲ Missed gains if not positioned
▼ Realized losses if not hedged
No downside captured in this branch's estimates.
Magnitudes assume — IF the branch materialises — the moves described. Actual moves depend on timing, prior positioning, and intervening events.
Policy lens —FERC opens an emergency rulemaking on data-centre power interconnection priority; the Commerce Department publishes updated AI Export Administration Regulations restricting advanced-node GPU exports; the EU AI Office activates its first systemic-risk designation under the AI Act.
Trade lens —NVDA and ASML sustain pricing power; data-center and nuclear-PPA names bid on baseload scarcity; INTC AI-chip share compresses. · meaningful · slow
Outcomes below — each % shown is the overall probability of that full chain occurring
If this path occurs — possible outcomes
Outcome % = conditional on this path occurring · Path % = joint probability of this exact chain from today
Policy lens —The FTC opens a Section 2 monopolisation investigation into Nvidia's CUDA and packaging ecosystem; the Commerce Department issues updated Validated End-User authorisation requirements for advanced GPU exports; the EU Digital Markets Act Directorate opens a gatekeeper designation proceeding for Nvidia.
Trade lens —NVDA, TSM and ASML compound the order-book moat; challengers lose datacenter share; CoWoS capacity remains the binding constraint. · meaningful · slow
Policy lens —The Commerce Department approves AMD and Trainium export licences for allied-country datacentres under a revised Validated End-User framework; Congress holds hearings on AI chip competition and supply-chain diversity; the EU AI Office updates its GPAI model-provider register to include Amazon and Google custom silicon.
Trade lens —AMZN and GOOG custom silicon re-rate on AI-chip share gain; NVDA margins compress toward 65-70%; SK Hynix HBM demand diversifies. · meaningful · slow
Policy lens —FERC rescinds priority interconnection status for speculative datacentre projects; the Commerce Department reassesses export-licence backlogs as demand signals weaken; state-level PUCs in Texas and Virginia open rate-case reviews on datacentre power agreements.
Trade lens —NVDA -40% on inventory correction; data-center and PPA-demand assumptions cut; MSFT and GOOG application-layer multiples expand on cheaper infra. · structural · slow
Information cutoff: 2026-05-21 · Authored: AI-generated, council-reviewed · Live signal counts updated hourly