Restrictions on cross-border flow of frontier AI compute (GPUs, accelerators) and model weights. Distinct from `compute_reporting` (which is disclosure) — this is restriction of access by recipient. US BIS rules (Oct 2023 advanced computing, Jan 2025 outbound investment), EU dual-use Regulation 2021/821 overlay, China retaliatory measures + indigenisation push. Mostly outside traditional AI-governance instruments; carving its own track.
Definition & scope
The cross-jurisdiction picture below shows how each of 45 tracked instruments treats this topic. The patterns vary substantially — and 41 regimes are silent, leaving gaps that future policy work could address.
Regulatory approaches
The table records only verdicts, not the legal machinery. The dominant instrument is the US Export Administration Regulations (EAR): the October 2023 BIS rules control advanced chips via performance-defined classifications — ECCN 3A090/4A090, the .a tier capturing ICs at total-processing-performance ≥ 4800 (BIS, 88 FR, doc. 2023-23055). That compute itself is the lever reflects an argument it is uniquely governable because "detectable, excludable, and quantifiable" and made via a concentrated supply chain 1. Whether such restrictions achieve their economic aim is contested empirically: a non-Western quantitative assessment estimates measurable but bounded effects on Chinese semiconductor activity 2. January 2025's "Framework for AI Diffusion" extended this to intangibles, creating ECCN 4E091 for closed-weight models trained on >10^26 operations (BIS, 90 FR, doc. 2025-00636). The EU differs: Regulation (EU) 2021/821 lets member states control unlisted items (Art. 9) via end-use catch-all provisions (Arts. 4-5). China relies on its Export Control Law (2020) plus MOFCOM measures, not any AI-specific statute (CSIS 2024).
Key fault lines
Three contested axes structure the debate. First, **unilateral vs plurilateral.** Leading-edge lithography is a single-firm chokepoint (ASML), so US measures depend on Dutch/Japanese alignment; "weaponized interdependence" holds chokepoint leverage erodes when allies decline 3. The 2022 controls were read as weaponizing US dominance over the value chain (Allen 2022, csis.org); others argue the strategy is "increasingly proving to be a fallacy" as Chinese firms circumvent it 4. Second, **the object — chips, equipment, or weights.** ECCN 4E091's bid to control closed-model weights was contested as unworkable for an exfiltrable intangible, rescinded pre-effect. Even for chips, smuggling "is already happening to a limited extent" and may grow (Grunewald 2023, iaps.ai), prompting firmware-based licensing proposals 5. Alongside state export controls, frontier labs now self-impose weight-flow governance: Anthropic's Responsible Scaling Policy ties ASL-3+ tiers to model-weight access controls (a recipient-restriction analog) (Anthropic 2024), DeepMind's Frontier Safety Framework adds weight-access mitigations and restricted-deployment options, and Meta's Frontier AI Framework's release decisions implicitly determine cross-border weight flow.
Trajectory — what is changing
Among the most volatile areas in AI governance; verdicts are snapshots. **Oct 2023** — BIS tightens advanced-computing controls (ECCN 3A090/4A090) and closes 2022 loopholes (BIS, doc. 2023-23055). **Dec 2024** — China bans gallium, germanium, antimony exports to the US as a countermeasure (PRC MOFCOM Notice 2024 No. 46). **Jan 2025** — BIS issues the AI Diffusion Framework with ECCN 4E091 weight control (BIS, doc. 2025-00636). **May 13 2025** — BIS rescinds it two days before the May 15 effective date, citing bureaucratic and diplomatic harm (BIS 2025). A key caution in the empirical literature is that restriction can backfire: sanctioned Chinese firms raised R&D ~49% and patenting ~41% under controls 6, while estimates of the controls' actual economic bite on Chinese chipmaking remain modest 2, amid intensifying East-Asian catch-up 7. The long-promised Diffusion replacement remains unsettled as of mid-2026.
Coverage across jurisdictions
Historical primacy & cross-jurisdiction tension
First addressed by Executive Order 14110 on Safe, Secure, Trustworthy AI on (implicit). Subsequent regimes have either codified, diverged from, or remained silent on this baseline.
Compare jurisdictions: EU vs US · EU vs UK · EU vs CN
Enforcement & impact
Silent regimes — gap signal
Instruments that do not address Compute + Model-Weight Export Controls — candidates for future policy work.
- EU AI ActEU
- Executive Order 14179 — Removing Barriers to American Leadership in AIUS
- UK Pro-Innovation Approach to AI Regulation (White Paper)UK
- Interim Measures for Generative AI Service ManagementCN
- G7 Hiroshima AI Process Code of ConductG7
- OECD AI Principles (Recommendation)OECD
- Council of Europe Framework Convention on AIcouncil_of_europe
- UN GA Resolution on Safe, Secure, Trustworthy AIUN
- NIST AI Risk Management FrameworkUS
- Bletchley Declaration on AI Safetyglobal
- Seoul Declaration on Safe, Innovative and Inclusive AIglobal
- NIST AI RMF Generative AI ProfileUS
- California SB-1047: Safe and Secure Innovation for Frontier AI Models ActUS
- India Digital Personal Data Protection Act + AI Advisory (MEITY)IN
- Brazil AI Bill (PL 2338/2023)BR
- ASEAN Guide on AI Governance and EthicsASEAN
- African Union Continental AI StrategyAfrican_Union
- OpenAI Preparedness FrameworkUS
- UK-US AI Safety Institute Memorandum of Understandingglobal
- White House Voluntary AI CommitmentsUS
- Singapore Model AI Governance Framework for Generative AISG
- Japan METI AI Guidelines for BusinessJP
- General Data Protection Regulation (GDPR)EU
- EU General-Purpose AI Code of PracticeEU
- OMB Memorandum M-24-10 (Advancing Governance, Innovation, and Risk Management for Agency Use of AI)US
- GSA Generative AI and Specialized Computing Infrastructure Acquisition Resource GuideUS
- DoD Responsible AI Strategy and Implementation PathwayUS
- FedRAMP AI Cloud Procurement GuidanceUS
- DFARS Subpart 252.204 (Safeguarding Covered Defense Information and Cyber Incident Reporting)US
- California SB-53: Transparency in Frontier Artificial Intelligence Act (TFAIA)US
- California SB 243: Companion ChatbotsUS
- California SB 942: AI Transparency ActUS
- Revised Product Liability Directive (Directive (EU) 2024/2853)EU
- UNESCO Recommendation on the Ethics of Artificial IntelligenceUNESCO
- Directive (EU) 2024/2831 on improving working conditions in platform workEU
- Provisions on the Administration of Deep Synthesis of Internet Information ServicesCN
- New York RAISE Act: Responsible AI Safety and Education ActUS
- TAKE IT DOWN Act (Tools to Address Known Exploitation by Immobilizing Technological Deepfakes on Websites and Networks Act)US
- Italy Law No. 132/2025 on Artificial Intelligence (Legge 23 settembre 2025, n. 132)IT
- Japan AI Promotion Act (Act on the Promotion of Research, Development and Utilization of AI-Related Technologies)JP
- UN Global Digital CompactUN
Further reading
9 academic & grey-literature sources bearing on this topic — catalogued metadata with a primary link; one-line findings are ✦ AI-generated summaries, labeled as such (charter §7.9). Browse the full literature index.
- China's semiconductor conundrum: understanding US export controls and their efficacy Peer-reviewed✦ AIArgues "America's chokepoint strategy is increasingly proving to be a fallacy": Chinese chipmakers have "managed to circumvent these measures" in four ways, accelerating domestic innovation.
- Export Controls and Innovation in Sanctioned Countries Working paper✦ AIUsing the 2007 US 'China Rule', finds sanctioned Chinese firms raised R&D by ~49% and patenting by ~41% — evidence export controls can accelerate the target's indigenous innovation.
- Computing Power and the Governance of Artificial Intelligence Preprint✦ AIArgues compute is a uniquely governable lever because it is "detectable, excludable, and quantifiable, and is produced via an extremely concentrated supply chain".
- Near-Term Enforcement of AI Chip Export Controls Using a Firmware-Based Design for Offline Licensing Preprint✦ AIProposes firmware 'disabling AI chips unless they have an unused license from a regulator', a hardware-enforceable mechanism for export-control compliance on chips like the H100.
- Geopolitics and the changing landscape of global value chains and competition in the global semiconductor industry: Rivalry and catch-up in chip manufacturing in East Asia Peer-reviewed✦ AIAnalyses how geopolitics reshapes semiconductor global value chains and East-Asian rivalry/catch-up, the structural backdrop against which chip export controls operate.
- AI Chip Smuggling into China: Potential Paths, Quantities, and Countermeasures Research institute✦ AIFinds AI chip smuggling into China "is already happening to a limited extent and may involve greater quantities in the future," proposing six countermeasures including a BIS chip registry.
- A Study on the Economic Effects of U.S. Export Controls on Semiconductors to China Peer-reviewed✦ AIEmpirically estimates the economic effects of US semiconductor export controls on China, a non-Western quantitative assessment of control efficacy.
- Choking Off China's Access to the Future of AI Think tank✦ AIAnalyzes the Oct 2022 controls as "weaponizing its dominant chokepoint positions in the global semiconductor value chain" to block China's access to AI chips, design software, and equipment.
- Weaponized Interdependence: How Global Economic Networks Shape State Coercion Peer-reviewed✦ AIThe 'chokepoint' and 'panopticon' theory of how states exploit central network hubs for coercion — the IR foundation for using concentrated chip supply chains as export-control leverage.
References
Sources cited inline in the analysis (linked from the superscript markers), then the primary instrument sources behind the classifications.
- Sastry, Heim, Belfield, Anderljung, Brundage, et al. (2024) Computing Power and the Governance of Artificial Intelligence, arXiv. arXiv:2402.08797 — Argues compute is a uniquely governable lever because it is "detectable, excludable, and quantifiable, and is produced via an extremely concentrated supply chain". ↩
- Do-Joon Park, Shuzhi Liu (2023) A Study on the Economic Effects of U.S. Export Controls on Semiconductors to China, International Commerce and Information Review (Korea Interna. 10.16980/jitc.19.1.202302.129 — Empirically estimates the economic effects of US semiconductor export controls on China, a non-Western quantitative assessment of control efficacy. ↩
- Henry Farrell, Abraham L. Newman (2019) Weaponized Interdependence: How Global Economic Networks Shape State Coercion, International Security. 10.1162/isec_a_00351 — The 'chokepoint' and 'panopticon' theory of how states exploit central network hubs for coercion — the IR foundation for using concentrated chip supply chains as export-control leverage. ↩
- Megha Shrivastava and Amrita Jash (2025) China's semiconductor conundrum: understanding US export controls and their efficacy, Cogent Social Sciences. 10.1080/23311886.2025.2528450 — Argues "America's chokepoint strategy is increasingly proving to be a fallacy": Chinese chipmakers have "managed to circumvent these measures" in four ways, accelerating domestic innovation. ↩
- James Petrie (2024) Near-Term Enforcement of AI Chip Export Controls Using a Firmware-Based Design for Offline Licensing, arXiv (cs.CR). arXiv:2404.18308 — Proposes firmware 'disabling AI chips unless they have an unused license from a regulator', a hardware-enforceable mechanism for export-control compliance on chips like the H100. ↩
- Xueyue Liu, Yu Liu, Alexey Makarin, Jaya Wen (2025) Export Controls and Innovation in Sanctioned Countries, Harvard Business School Working Paper 25-004. source — Using the 2007 US 'China Rule', finds sanctioned Chinese firms raised R&D by ~49% and patenting by ~41% — evidence export controls can accelerate the target's indigenous innovation. ↩
- Chan-Yuan Wong, Henry Wai-chung Yeung, Shaopeng Huang, Jaeyong Song, Keun Lee (2024) Geopolitics and the changing landscape of global value chains and competition in the global semiconductor industry: Rivalry and catch-up in chip manufacturing in East Asia, Technological Forecasting and Social Change. 10.1016/j.techfore.2024.123749 — Analyses how geopolitics reshapes semiconductor global value chains and East-Asian rivalry/catch-up, the structural backdrop against which chip export controls operate. ↩
- US-EO-14110: §4.2(b) directs export-control coordination via BIS; not the primary venue but the policy hook
- ANTHROPIC-RSP-2024: ASL-3+ tiers include model-weight access controls (recipient-restriction analog)
- DEEPMIND-FSF-2024: FSF mitigations include model-weight access controls + restricted-deployment options
- META-FRONTIER-2024: Framework's release decisions implicitly determine cross-border weight flow
Cite this article 8 formats · BibTeX, RIS, APA, Chicago, … · 1-click copy
Persistent identifier: https://policywindow.org/wiki/compute-export-controls — committed-stable URL with content-versioning via ?asOf= (rollout pending per methodology §7). DOIs via Zenodo are on the roadmap.
Article tools — track changes, suggest an edit
View history — every captured revision of this article · What links here
4 instruments tracked.
Does governance work? — the social-science evidence
What the peer-reviewed social science shows: whether the harm this topic addresses is empirically real, and whether governance of it works. The badge is the epistemic status of the evidence(not the policy debate) — “thin” or “absent” efficacy evidence is itself a finding (the “second silence”). Each epistemic-status label is Policy Window's editorial assessment of the cited evidence base (a structured classification), not a verdict any single source issues.
That compute controls materially constrain China's frontier-AI hardware access is empirically real and measured: by mid-2025 the US hosted ~75% of catalogued AI-supercomputer performance versus China's ~15% (Pilz, Sanders, Rahman & Heim 2025), corroborated by the Federal Reserve's estimate of ~74% US vs ~14% China high-end AI compute share (Haag, FEDS Notes, Oct 2025), and US prosecutions document large diversion networks (e.g. the ~$160M Alan Hao Hsu / Hao Global H100/H200 case — the first 'AI diversion' conviction, guilty plea Oct 2025, SDTX). The honest caveat is that the magnitude of the binding constraint is genuinely contested: DeepSeek's V3/R1 reached near-frontier capability at an order of magnitude less reported compute via algorithmic efficiency, and analysts argue the controls have simultaneously accelerated Chinese state-backed indigenization, so whether the controls slow capability (the actual aim) versus merely shift its cost structure remains unsettled.
Sources: Pilz, Sanders, Rahman & Heim 2025 (Trends in AI Supercomputers, arXiv:2504.16026 / Epoch AI) — VERIFIED, gives US ~75% / China ~15%; Haag 2025 (FEDS Notes, 'The State of AI Competition in Advanced Economies', Federal Reserve, 6 Oct 2025) — VERIFIED, gives US 74% / China 14% / EU 4.8% of high-end AI compute; US v. Alan Hao Hsu / Hao Global 2025 (DOJ, SDTX; ~$160M H100/H200 diversion, guilty plea Oct 2025, first AI-diversion conviction) — VERIFIED
There is no rigorous impact evaluation showing that compute or model-weight export controls achieve their stated strategic aim of durably slowing frontier-AI capability diffusion to China — the regime is too recent, the counterfactual is unidentified, and the most ambitious instrument (the Jan 2025 BIS 'Framework for AI Diffusion', ECCN 4E091 covering model weights of closed models trained on >10^26 operations) was rescinded on 12-13 May 2025 before it ever took effect (its enforcement date was 15 May 2025), so the evidence that the rule works is itself missing. The closest analogue evidence base, the economic-sanctions evaluation literature, is sobering: Hufbauer, Schott, Elliott & Oegg (2007) coded roughly a third of their historical cases as 'successful' (their database covers ~170-200 cases since WWI; the disputed coding was ~40 of 115, ~34%), but Pape's reanalysis (1997/1998) argued the genuinely sanctions-attributable success rate was far lower (he recoded it to ~5 of 115, under 5%), and the broader literature finds efficacy decays as targets adapt and substitute — the precise dynamic export-control critics attribute to Chinese indigenization and smuggling. This is an analogue, not direct evidence on export controls.
Sources: Hufbauer, Schott, Elliott & Oegg 2007 (Economic Sanctions Reconsidered, 3rd ed., Peterson Institute for International Economics) — VERIFIED; Pape 1997/1998 (Why Economic Sanctions Do Not Work, International Security 22(2), 1997; Why Economic Sanctions Still Do Not Work, International Security 23(1), 1998) — VERIFIED; BIS 2025 (Framework for Artificial Intelligence Diffusion, Federal Register doc 2025-00636 / 90 FR, eff. 13 Jan 2025; ECCN 4E091 model-weight control; rescinded 12-13 May 2025 before its 15 May effective date) — VERIFIED