Radar · Beta
Who supplies AI compute, where the physical chokepoints are, and what's shifting in the fab, datacentre and power layers beneath the model layer.
incumbents vs new entrants · ranked by mentions in compute/infrastructure coverage
from the compute/infrastructure-tagged set, most recent first
Higher-bandwidth optical interconnects can support the scaling of AI clusters, making networking capacity an increasingly important constraint and investment area in AI compute infrastructure.
fab capacity · HBM · packaging · grid · export controls
Greater availability of a low-latency inference accelerator could support more responsive real-time and agentic AI deployments.
Large-scale AI deployment may increasingly depend on capital-market capacity to fund hardware and data-centre expansion, not only on chip supply.
Coming soon
A timeline of announced compute capacity commitments — new fab, datacentre and cluster buildouts, tracked by announcement date.
Waiting for: A larger population of items carrying both a computeCategory and a verified announcement date; the fields exist but are not live yet (migration not applied).
Coming soon
Announced AI-related investment broken down by compute category — accelerators, datacentres, fabs, networking, energy, memory, cloud capacity.
Waiting for: The AI desk's own investment extraction (FR-613, shipped this sprint) to accumulate enough rows to chart — today's 155 investments rows are still almost entirely defence-desk.
Coming soon
The split between compute/infrastructure coverage that serves training workloads versus inference/deployment workloads, over time.
Waiting for: workloadPhase (FR-614, shipped this sprint) to be live and populated — its migration is written but not applied, so no item carries it yet.
Coming soon
Which countries are building, exporting, or restricting AI compute capacity — a measured pillar, not a hand-curated one.
Waiting for: Enough geo-tagged compute/infrastructure coverage to measure a posture per country rather than curate one — deliberately deferred rather than shipped as a guess (see the roadmap's excluded-scope note).
If realised, the spending could accelerate global AI compute and networking capacity while increasing the strategic importance of suppliers of advanced data-centre interconnect equipment.
Memory availability and pricing are becoming material constraints on AI infrastructure expansion, potentially increasing the cost and supply-chain dependence of cloud and compute operators.
Higher-bandwidth, lower-latency interconnects could support larger and more communication-intensive AI clusters, but the forecast is not independently validated in the source.
If sustained, higher inference costs and weak consumer monetisation could influence semiconductor, server and deployment strategies, although the source provides limited evidence beyond forum commentary.
More efficient chip-to-data-centre links could support continued scaling of AI systems and alter the infrastructure dependencies of advanced AI deployment.
If accurate, the capacity gap could constrain Taiwan's ability to support domestic AI inference workloads and weaken its position relative to South Korea in AI infrastructure.
More efficient use of AI chips could ease physical and economic scaling constraints, but the supplied text does not provide specific projects, technologies, performance data, or commitments.
The update indicates that AI infrastructure demand is expanding beyond compute components into thermal management and backup-power systems, although it does not document a specific capacity increase or deployed capability.
Greater concentration of Nvidia-related AI memory demand at SK Hynix could affect supplier dependence, competitive positioning, and the resilience of the AI hardware supply chain.
Production availability of a specialised inference accelerator could improve the responsiveness and deployment economics of agentic AI, although the source provides no independent performance measurements or production-volume details.
Faster token generation can improve the responsiveness and practical deployment of agentic AI systems, although the source provides no independent performance evidence.
Higher inference efficiency could reduce the power and infrastructure required to operate increasingly token-intensive AI agents.
If implemented as described, the deployment could add purpose-built CPU capacity for agentic AI workloads, although the source does not independently substantiate the scale or expected gains.
More efficient and compact power delivery could reduce energy use, cost and space requirements in AI data-centre systems.
