Start here
Read the cornerstone Maps first, then open Dossiers for the actors controlling the mechanism.
Topic Hub
AI, compute, and geopolitics.
This hub tracks how compute access, advanced chips, capital allocation, coalitions, and state action shift AI capability. It is a curated operating picture, not a feed.
Evidence rule
Treat SOURCED claims as the floor. Follow INFERRED claims only when citations and weak points are visible on the page.
Quarterly cadence
This hub is revised on a quarterly authority pass, with earlier changes folded into canonical Map changelogs.
Compute
Training access, cloud concentration, data centers, energy constraints, and capability timelines.
Chips
Export controls, packaging, interconnects, foundry capacity, and semiconductor chokepoints.
Capital
Subsidies, capex cycles, procurement, venture pressure, and state-backed financing.
Coalitions
Allied controls, standards bodies, enforcement coordination, and diplomatic alignment.
Evidence and category boundary
This hub is for AI power shifts explained through compute, chips, capital, coalitions, infrastructure, export controls, and state action. It intentionally excludes generic product launches, thin recap pages, and unsourced capability claims.
Working definitions for AI statecraft
Hardware geofencing
A control pattern that restricts the activation, service, or performance of hardware systems based on jurisdiction, location, or approved deployment environment.
Compute integrity
The verifiable condition that a compute cluster is operating with approved hardware, software, telemetry, and policy controls intact.
Remote disablement
A hardware or platform capability that can revoke service, access, or cryptographic trust when a regulated condition is violated.
Cluster telemetry
Operational logs and measurements used to prove workload type, utilization, attestation state, topology, and policy compliance across a compute environment.
Infrastructure licensing
A regulatory model that governs access to advanced compute, cloud infrastructure, or semiconductor systems through formal authorization conditions.
Attestation clause
A legal or policy requirement that obliges an operator or vendor to prove the state, origin, or compliance condition of a system.
Canonical Maps
Top authority pages
JEDEC finalized the JESD270-4 HBM4 standard (April 21, 2026), doubling the interface width to 2048-bit and targeting 2 TB/s bandwidth.
DeepSeek V4, a 1-Trillion parameter MoE model, was released on April 24, 2026.
March 20, 2026: Huawei officially launches the Ascend 950PR (Atlas 350) AI accelerator, delivering 1.56 PFLOPS (FP4) — roughly 2.8x the performance of Nvidia's compliant H20 card.
The US Department of Commerce drafted sweeping 'Global Permit' regulations in March 2026 to govern AI chip exports to virtually every nation.
Advanced packaging throughput now materially shapes deployment timelines for high-end AI accelerators.
Cloud compute screening rules expand
NVIDIA positioned Blackwell Ultra as the next Blackwell platform, centered on GB300 NVL72 and HGX B300 NVL16 for reasoning, agentic AI, and test-time scaling; NVIDIA said GB300 NVL72 delivers 1.5x more AI performance than GB200 NVL72.
Canada officially opened the application intake for its five-year AI Sovereign Compute Infrastructure Program (AISCIP) to establish Canadian-owned and operated AI supercomputing facilities.
Construction for Micron's ¥1.5 trillion ($9.6 billion) DRAM plant expansion in Higashihiroshima, Japan begins in May 2026, aimed at high-volume advanced HBM production and localizing advanced packaging within the G7 security perimeter.
The Trump-Xi Beijing Summit (May 14-15, 2026) concluded without a formal agreement on AI chip export relaxation, leaving Nvidia H200 deliveries stalled despite warm rhetoric.
Power Lens Spine
Grouped for human and machine scanning
Compute
Training access, cloud concentration, data centers, energy constraints, and capability timelines.
Chips
Export controls, packaging, interconnects, foundry capacity, and semiconductor chokepoints.
Capital
Subsidies, capex cycles, procurement, venture pressure, and state-backed financing.
Coalitions
Allied controls, standards bodies, enforcement coordination, and diplomatic alignment.
Priority Dossiers
Actors and institutions with durable leverage
Dominant provider of accelerated computing infrastructure and architect of Sovereign AI ecosystems. • United States / Global
Advanced semiconductor foundry and AI compute supply-chain chokepoint • Taiwan
Critical lithography supplier; focal point of G7 semiconductor alignment tensions following the Dutch government's May 14, 2026 explicit parliamentary objections to unilateral US export control expansions (MATCH Act). • Netherlands / Global
China foundry anchor in domestic semiconductor manufacturing capacity and industrial policy execution. • China
China technology platform company with strategic influence across telecom, cloud, and AI hardware/software ecosystems. Primary global alternative to Western AI compute stacks via the Ascend NPU line, self-developed HiBL memory, and AgentArts agentic OS. • China / Global
Frontier AI lab and infrastructure operator shaping gigawatt-scale data center deployment, advanced networking systems, and U.S. government adoption pathways. • United States / Global
Frontier AI lab securing hyperscale, specialized-cloud, and state-facing alliances to expand Claude deployment and compute access. • United States / Australia / Global
Global leader in frontier AI labs, hyperscale cloud, and policy enforcement infrastructure with strong capital and standards-setting power. • United States
Major AI and semiconductor system competitor with strong domestic market scale, industrial policy coordination, and cloud/platform depth. • China
Critical global semiconductor manufacturing hub with outsized influence over advanced node output and packaging ecosystem resilience. • Taiwan
Global cloud and AI infrastructure operator using sovereign partitions, custom silicon, and frontier-lab capacity agreements to shape where advanced AI can train, deploy, and stay under jurisdictional control. • United States / Europe / Global
Platform-scale AI infrastructure operator scaling custom silicon, cloud CPUs, power procurement, and hyperscale data-center buildout. • United States / Global
Quarterly authority review
Every quarter, this hub is reviewed for duplicate story arcs, orphaned dossiers, weak source stacks, and Maps that no longer deserve indexable status. The goal is fewer pages with stronger receipts, not higher volume.
Technical-policy convergence
The site should be readable by both compliance counsel and systems operators. Translatory briefs are the bridge: they pair the legal clause with the control surface and the telemetry path that would prove a claim in practice.
Mandate to mechanism
State the exact clause, draft, filing, or official mandate that creates the obligation.
Describe the silicon, firmware, platform, cloud, or network surface where the obligation is enforced.
List the telemetry, attestation, procurement, or inspection path that would prove the control is real.
Policy to failure mode
Name the enforcement trigger, threshold, or prohibited condition in legal language.
Map that trigger to the system component that can throttle, deny, log, or disable behavior.
Explain what would falsify the implementation claim or reveal a compliance gap.
Operator implications
Identify which operator, vendor, or jurisdiction is bound by the obligation.
Translate the obligation into operational burdens such as monitoring, capacity planning, key management, or hardware replacement.
Show which records, audit artifacts, or public statements would confirm those burdens in practice.