Executive Summary
Generated July 30, 2026
Data Center Infrastructure
Data Center Power & Cooling
Strategic importance is our editorial rating of how central this entity is to the technology landscape we track; confidence reflects how well-sourced and current the underlying evidence is.
3
Chokepoint Score
3 other entities structurally depend on this one
Overview
Data center power and cooling technology has become one of the central bottlenecks in AI infrastructure buildout, as GPU rack densities outpace what traditional air cooling can handle. The data center liquid cooling market is projected to grow from $4.07 billion in 2026 to $27.65 billion by 2033 (a 31.5% CAGR) as liquid cooling goes mainstream across the dozens of gigawatt-scale AI factories under construction globally.
Strategic Connections
- Constellation Energy: Constellation Energy supplies power generation capacity, including nuclear, to meet the electricity demands of AI data centers.
- Oklo: Oklo is developing small modular nuclear reactors specifically to provide dedicated, behind-the-meter power for AI data centers, including a 1.2GW campus for Meta.
- NuScale Power: NuScale Power's NRC-certified small modular reactors are being positioned as dedicated power sources for AI data centers, including a Standard Power Ohio deal.
- Kairos Power: Kairos Power's advanced reactors are being developed to deliver up to 500MW of dedicated carbon-free power for Google's AI data centers by 2035.
- GE Vernova: GE Vernova's heavy-duty gas turbines, sold out through 2030-2031, are a primary bottleneck and enabler for AI data center power buildout.
+22 more connections in the Full Analyst Report.
Strategic Risks
- Electricity supply constraints, not just cooling technology, are increasingly the binding limit on AI data center buildout speed
- Rapid transition to liquid cooling requires significant retrofit or new-build capital, creating execution risk for operators moving too slowly
Future Outlook
- Liquid cooling expected to become the default (not niche) approach for new AI data centers within the next few years
- Direct power-purchase and generation deals with utilities and nuclear operators expected to keep expanding as electricity, not chips, becomes the binding constraint
Sources
2 sources, 57% average source confidence — full citations in the Full Analyst Report.
Generated by InsightNodes — Technology Intelligence Platform. This report reflects sourced, evidence-backed information as of the dates cited above and is not investment advice.
Viewing the Executive Summary — switch to the Full Analyst Report.