Data Centers: Deep Tech’s De-Risking Sandbox

The data center industry is entering a transformative era as artificial intelligence (AI) drives unprecedented energy demands. With advanced AI models requiring up to 1,000 times the power of their predecessors, (see estimates from Sustainable Metal Cloud below), the power consumption of data centers is expected to accelerate dramatically. This rapid growth poses challenges and opportunities for sustainability and operational efficiency, prompting industry leaders to explore innovative solutions to decarbonize these energy-intensive hubs. 

Uneven Energy Demands: A U.S. Case Study 

As data center power consumption grows, its impact is unevenly distributed. States with significant data center infrastructure, such as Virginia and Texas, are experiencing accelerated energy demand. Lawrence Berkeley National Lab highlights grid interconnection queues, with solar and energy storage projects delayed as regions scramble to meet growing needs. Yet, clean power supply remains insufficient, creating an urgent need for innovative energy solutions. 

Challenges at the Forefront 

1. Speed of Deployment: Technologies that enable rapid data center deployment to meet immediate AI demands.
2. Energy Optimization: Reducing operational energy consumption for both cost and environmental sustainability.
3. Long-Term Reliability: Ensuring data centers can integrate clean baseload power while mitigating risks from external factors like extreme weather. 

Accelerating Deployment: Modular Data Centers and Localized Energy 

Innovators are addressing the challenge of rapid deployment with modular data centers and by locating facilities near renewable energy sources. 

• Flexnode specializes in pre-fabricated liquid-cooled modular data centers. These high-density centers are designed for rapid deployment in extreme environments while minimizing water usage. Flexnode also collaborates with the Department of Energy’s ARPA-E Coolerchips program to advance cooling technologies.

• Crusoe Energy leverages “digital flare mitigation,” converting wasted natural gas into energy for AI-driven data centers. By situating data centers near generation sites, Crusoe reduces transmission losses and maximizes power efficiency. Just last week, Crusoe raised a $600M Growth Equity round to advance its low-carbon AI cloud, including financing from Peter Thiel’s Founders Fund.  

Cooling Innovation: Tackling a 40% Energy Burden 

Cooling accounts for 40% of a data center’s energy use, making it a critical focus for decarbonization. Traditional air-cooling systems struggle to meet the demands of high-density server racks, prompting the rise of liquid cooling solutions. 

• LiquidStack develops both immersion (single and two-phase options) and direct-to-chip cooling systems that reduce energy use by up to 95%. These systems are ideal for high-performance AI applications, minimizing both capital and operational costs.

• Submer provides single-phase immersion cooling systems, fully submerging servers in non-conductive liquids to optimize heat dissipation.

• Jetcool focuses on microconvective direct-to-chip cooling, using precise fluid jets to manage hotspots in AI systems. Their technology is highly effective for ultra-dense server environments. 

Driving Clean Baseload to Market: Enhanced Geothermal  

Renewable energy sources are essential to decarbonizing data centers, with geothermal and nuclear power leading the charge. 

• Fervo Energy adapts hydraulic fracturing techniques from oil and gas to extract heat from “hot dry rock.” Partnering with Google, Fervo powers a Nevada facility, supplying clean energy for Google data centers.

• Sage Geosystems pioneers geopressured geothermal systems, combining heat and pressure to generate consistent power. Their partnership with Meta aims to supply 150 MW of energy by 2027. 

On-site Power Management 

Optimizing data center energy usage goes beyond cooling. AI-enabled tools and advanced power electronics are transforming how energy is managed. 

• Phaidra develops AI-driven software to optimize cooling and energy consumption, achieving up to 30% energy reductions in pilot programs.

• Daanaa creates Power Transaction Units that streamline electricity conversions, reducing energy losses and minimizing heat generation. These systems integrate with solar, storage, and transmission infrastructures for maximum efficiency. 

 Grid Resilience Tech Can Tackle Data Center Downtime 

As AI becomes more critical in the global economy, the risk of downtime due to power disruptions is significant in both a financial and service delivery respect. Grid hardening technologies are being developed to address vulnerabilities in power transmission and distribution and will likely experience a pull-through effect from data center market growth.  

• TS Conductor designs carbon composite conductors that increase capacity and resilience, preventing thermal sag during high-heat events.

• Veir introduces high-temperature superconductors that cool wires internally, enhancing capacity while protecting against extreme weather. 

Transformers play a central role in ensuring stable energy distribution for data centers, particularly as power demand grows exponentially with AI-driven operations. As data centers become more energy-dense, innovative transformer technologies are emerging to optimize power flow and prevent inefficiencies. 

• Amperesand is developing solid-state transformers using semiconductors to convert / control power – highly flexible, allowing users to isolate loads and prevent faults. At scale, they have the potential to be a powerful solution for grid resilience against disasters and extreme weather. It offers efficiency and resilience to non-grid customers that have high degrees of power use variability.

• IONATE has re-invented the inner workings of transformers without re-inventing the form factor – providing a novel approach to transformers but still a drop-in. The company delivers highly precise electromagnetic power flow controls through an AI layer, that can modulate flow in real-time and at a high level of granularity. The end result is the addition of 33% more distributed energy resources connected to the grid and 20% more capacity through poles and wires. 

Looking Ahead 

The data center industry is at a pivotal moment, with hyperscalers driving innovation and investment in clean technologies. From modular designs and advanced cooling systems to renewable energy integration, the ecosystem is evolving rapidly to meet AI’s energy demands sustainably. Technologies de-risked in data centers, such as advanced geothermal and SMRs, may have far-reaching impacts across industries, from grid optimization to electric vehicles. 

We predict that technology de-risked across the data centers and AI infrastructure continuum will be a key level of activity in 2025.  Keep an eye out for future analyses on these topics.