Cascade Simulation — Disruption Propagation Through the AI Compute Supply Chain

Gringras, David

Scrutica

Graph written 2026-10-24

Cascade

Simulate disruptions through 11,920 supplier–customer relationships among 4,201 organizations

5 corrections · latest 23 Sep 2026

The cascade engine

Scenario definitions

Downstream connections. Organizations ranked by the number of others connected downstream within 3 recorded supplier–customer links, each counted once. The count records relationships and says nothing about volume, share or substitutes.

This count uses the publicly inspectable relationships. The simulation also uses licensed relationships, weights each by the evidence for it and stops at an impact cutoff, so its results are not read off this curve.

Relationships recorded as ended, or last disclosed five or more years before the latest filing wave, are not drawn; an undated relationship can still name a company since acquired or dissolved.

Open the cascade engine

Across this graph, an organization has an average of 1.6 downstream organizations within one link, 5 within two, and 12 within three. Within three links, none connects to 100 others downstream (the most connected, to 98); 172 have none recorded. These counts do not account for the strength of a dependency or the availability of substitutes. How propagation is derived

Graph coverage

11,920: relationships used in simulations across 4,201 organizations. Of these, 1,924 come from a licensed layer limited to relationships between organizations already in Scrutica's records.
997: publicly inspectable relationships across 587 organizations, drawn in the graph and used to calculate the connections curve. Additional licensed relationships are included in simulations but cannot be redistributed. The drawn graph’s JSON lists 2,625 relationships across 1,355 organizations: these 997, 627 partnerships, 5 relationships recorded as ended and 996 relationships whose latest disclosure covers a fiscal period ending on or before 2019-12-31. Only the 997 are drawn, and only they propagate: a partnership moves nothing from supplier to customer, a relationship recorded as ended moves nothing now, and a relationship with no disclosure in the five years to the latest filing wave is not evidence of a present dependency.
20,533: source records, including 8,613 the simulation does not propagate along: another source’s record of the same relationship, an unnamed counterparty aggregate, a partnership, a relationship recorded as ended, or one last disclosed five or more years before the latest filing wave.

The pre-built scenarios 20 seeds

Severity sets the initial impact at the selected companies; some presets use reported exposure as a proxy, while others assume a loss. Decay reduces impact at each supply-chain hop and is always an editorial assumption. The source tier describes the evidence informing the scenario, including company filings and market research.
Scenario	Category	Seed nodes	Severity	Tier	Decay
TSMC Taiwan Disruption	Natural disaster	TSMC	1.00	T4	0.90
ASML Export Halt	Export control	ASML	1.00	T4	0.95
US–China Full Decouple: Revenue Impact	Geopolitical	NVIDIA, AMD, Intel, Qualcomm Inc	0.15	T4	0.70
US–China Full Decouple: Access Denial	Geopolitical	SMIC, Huawei Technologies, Baidu, ByteDance, Alibaba, Tencent Holdings Ltd.	0.55	T4	0.80
CoWoS Advanced Packaging Bottleneck	Supply bottleneck	TSMC	0.40	T4	0.70
SK Hynix HBM Disruption	Supply bottleneck	SK Hynix	0.60	T4	0.80
EDA Export Restriction	Export control	Synopsys, Cadence	0.90	T4	0.80
Chiplet Assembly Disruption	Supply bottleneck	TSMC, Synopsys, Cadence	0.80	T4	0.75
Qatar helium disruption (March 2026)	Supply bottleneck	Samsung, SK Hynix, TSMC	0.65	T4	0.75
InP/EML Laser Shortage	Supply bottleneck	Coherent, Lumentum Holdings	0.85	T4	0.80
ABF build-up film supply (Ajinomoto)	Supply bottleneck	Ajinomoto Co., Inc.	0.90	T4	0.95
Gallium/germanium controls return (after 27 Nov 2026)	Geopolitical	Coherent, Skyworks Solutions, Qorvo, Wolfspeed, Infineon Technologies	0.70	T4	0.65
CoWoS 2026 Allocation Constraint	Supply bottleneck	TSMC	0.85	T4	0.85
Japan Semiconductor-Input Seismic Shock (Korea coupling)	Natural disaster	Shin-Etsu, SUMCO, Tokyo Electron, Advantest, Disco, Lasertec, Tokyo Ohka Kogyo, Resonac Holdings, Kioxia	0.70	T4	0.82
Allied Coordination: TSMC Advanced Node Restriction	Policy intervention	TSMC	1.00	T4	0.90
EDA License Restriction: China Entity List Expansion	Policy intervention	Synopsys, Cadence	0.95	T4	0.80
Netherlands DUV Export Restriction to China	Policy intervention	ASML	0.70	T4	0.85
China Critical Minerals Retaliatory Restriction	Policy intervention	Coherent, Skyworks Solutions, Qorvo, Wolfspeed, Infineon Technologies	0.70	T4	0.65
Multilateral Cloud Compute Access Restriction	Policy intervention	Amazon.com, Inc., Microsoft Corporation, Alphabet, Oracle Corporation	0.80	T4	0.75
Netherlands EUV Restriction: Allied Countries Only	Policy intervention	ASML	1.00	T4	0.95

Related research

Epoch’s analysis of the CoWoS packaging and HBM memory chokepoint documents the concentration at both ends of the supply relationship: NVIDIA, Google, AMD and Amazon on the demand side; TSMC, SK Hynix, Samsung and Micron on the supply side. Relationships of that kind are the edges the simulator above propagates over.