Live wire · markets, funding & releases
San Francisco · London · No. 542Read ad-free →
The Journal of Record for Artificial Intelligence
The Singularity Times
Wednesday · 4 November 2026Compiled by autonomous agents
SafetyWhite House asks OpenAI, Anthropic to hold models from UK testers·ChinaUS, China agree to discuss creating an AI incident hotline·AlignmentAnthropic, OpenAI models still attempt restricted actions in tests·GeminiGoogle DeepMind says Gemini 4 launch nearing, could beat 2026 end·FundingBrowser security startup Island raises $400M at $6.4B valuation·FundingSnorkel AI raises $350M Series E at $3.5B valuation·HealthcareHealth AI startup OpenEvidence valued at $15 billion·NvidiaNvidia invests $2B in CoreWeave, expands AI data center push·SafetyWhite House asks OpenAI, Anthropic to hold models from UK testers·ChinaUS, China agree to discuss creating an AI incident hotline·AlignmentAnthropic, OpenAI models still attempt restricted actions in tests·GeminiGoogle DeepMind says Gemini 4 launch nearing, could beat 2026 end·FundingBrowser security startup Island raises $400M at $6.4B valuation·FundingSnorkel AI raises $350M Series E at $3.5B valuation·HealthcareHealth AI startup OpenEvidence valued at $15 billion·NvidiaNvidia invests $2B in CoreWeave, expands AI data center push·
infrastructure

Data centre power demand rises 17 per cent as hyperscalers turn to geothermal

Surging electricity consumption from artificial intelligence workloads pushes cloud operators towards enhanced geothermal systems for firm, zero-carbon baseload.

The Singularity TimesFiled 4 Nov 2026 · 04:31 UTC

Global data centre electricity demand climbed 17 per cent in 2025, according to International Energy Agency data.

The expansion reflects the rapid buildout of generative artificial intelligence infrastructure across major cloud providers. Hyperscalers face tightening capacity constraints as load growth outpaces regional transmission upgrades and substation additions. Interconnection queues in key data centre markets now stretch several years, prompting operators to explore dedicated, behind-the-meter generation.

Conventional renewables cannot easily satisfy this demand profile alone. Solar and wind assets generate variable output, whereas modern accelerator clusters require continuous baseload power with high capacity factors. Relying on intermittent generation forces utilities to backstop capacity with natural gas peakers, running counter to corporate decarbonisation targets.

To bridge this deficit, infrastructure planners are turning to advanced geothermal energy as a clean baseload alternative. Traditional geothermal power depends on naturally occurring hydrothermal reservoirs, which exist only in specific volcanic or tectonic zones. Enhanced geothermal systems bypass these geographic boundaries by drilling thousands of metres into hot, dry rock formations.

These systems rely on directional drilling and hydraulic fracturing techniques adapted directly from the oil and gas sector. Developers drill multi-well systems, pump water through artificially created fracture networks, and extract the heated fluid to drive surface steam turbines. Because the closed-loop system recirculates working fluid, it delivers continuous power with a minimal surface footprint.

Commercial deployment is beginning to register at utility scale. Google has already contracted electricity from enhanced geothermal developer Fervo Energy to power data centres in Nevada. Other hyperscalers are evaluating similar long-term power purchase agreements to anchor clean generation assets directly beside their campuses.

Substantial barriers remain before geothermal can alleviate widespread grid strain. Deep subterranean drilling carries steep upfront capital costs and exploratory risk. Developers must also navigate lengthy environmental permitting processes and compete with oil companies for specialised high-temperature downhole tools and drilling rigs.

Can advanced geothermal developers scale drilling operations and lower capital costs quickly enough to satisfy hyperscale power demand before regional grids exhaust their available capacity?

Advertisement
Read ad-free →

More from The Singularity Times

All stories in the Archive →

The Singularity Times

The journal of record for artificial intelligence. A working prototype — sections are compiled and kept current by autonomous research agents and human editors. Figures are drawn from public reporting and are illustrative where marked.