Aikido Technologies, a California-based floating wind power developer, has unveiled a bold plan to embed data centers within the underwater ballast tanks of its offshore wind turbine platforms. The startup aims to integrate 10–12 megawatts (MW) of AI compute capacity alongside 15–18 MW turbines and battery storage. A 100-kilowatt prototype is slated for testing off the coast of Norway by the end of 2026.
A New Frontier: Data Centers Beneath Offshore Wind Turbines
Aikido’s concept leverages the buoyancy tanks—ballasts—at the base of floating wind turbines. These tanks, filled mostly with fresh water, will house compact data halls of 3–4 MW each. The ocean serves as a natural cooling system, with heat transferred passively through steel walls into the surrounding seawater. The company assures that thermal impact will be limited to a few meters around the structure.
According to CEO Sam Kanner, “Before we go off‑world, we should go offshore.” He emphasizes that Aikido’s design combines proven offshore components with conventional data hall construction to create gigawatt-scale AI infrastructure that is faster, cleaner, cheaper, and more efficient than traditional land-based facilities.
Why It Matters: Sustainability Meets AI Demand
Traditional data centers are energy-intensive and strain power grids. In 2024, U.S. data centers consumed 183 terawatt-hours—about 4% of the nation’s total electricity. If growth continues unchecked, this could more than double by 2030.
Aikido’s model addresses both energy and cooling challenges. By co-locating compute with renewable generation and using seawater for cooling, the design reduces reliance on fossil fuels and freshwater resources. The integration of battery storage further enhances grid stability and operational flexibility.
Global Context: Underwater Data Centers Aren’t New
China’s HiCloud has already deployed a wind-powered underwater data center off Shanghai. The demonstration project delivers 2.3 MW of compute capacity and is part of a larger plan to scale to 500 MW. The facility, costing approximately $226 million, is cooled by seawater and powered almost entirely by offshore wind, achieving a power usage effectiveness (PUE) of 1.15—well below China’s 2025 mandate of 1.25.
HiCloud’s initiative underscores the viability of submerged data centers powered by renewables. However, challenges remain, particularly around servicing hardware and scaling operations.
Implications for Stakeholders
For the Tech Industry
- Energy Efficiency: Aikido’s design could dramatically reduce the carbon footprint of AI infrastructure.
- Grid Relief: Offshore deployment alleviates pressure on land-based power grids.
- Cooling Innovation: Passive ocean cooling offers a sustainable alternative to energy-intensive HVAC systems.
For Renewable Energy Developers
- New Revenue Streams: Wind farm operators could host compute infrastructure, adding value to their assets.
- Synergies: Co-locating generation and consumption reduces transmission losses and infrastructure costs.
For Regulators and Coastal Communities
- Environmental Impact: While Aikido claims minimal thermal disturbance, regulators will need to assess marine ecosystem effects.
- Permitting Complexity: Offshore installations face stringent regulatory scrutiny, especially in U.S. waters.
Challenges and Considerations
- Maintenance and Upgrades: Servicing submerged hardware could be costly and time-consuming.
- Scalability: Moving from a 100 kW prototype to gigawatt-scale infrastructure presents engineering and logistical hurdles.
- Environmental Oversight: Independent studies will be needed to validate claims about limited thermal impact.
Looking Ahead: What’s Next for Aikido?
Aikido plans to test its prototype off Norway’s coast by the end of 2026. If successful, the company envisions offshore wind farms capable of supporting 30 MW to over 1 GW of compute capacity. This could meet surging demand for AI infrastructure while minimizing environmental impact.
The concept aligns with broader trends in sustainable data infrastructure. In Japan, for instance, NYK Line is piloting offshore floating AI data centers powered by wind and cooled by seawater, targeting commercial operations by 2030.
Conclusion
Aikido Technologies is pioneering a novel approach to sustainable AI infrastructure by placing data centers beneath offshore wind turbines. This model promises to reduce energy consumption, leverage passive cooling, and integrate compute with renewable generation. While challenges around maintenance, scalability, and environmental impact remain, the concept represents a compelling step toward greener, more efficient data infrastructure. As the prototype moves toward testing in 2026, the tech and energy sectors will be watching closely.
Frequently Asked Questions
What is Aikido Technologies proposing?
Aikido plans to embed data centers inside the ballast tanks of floating offshore wind turbines, combining compute, renewable energy, and passive ocean cooling.
How much compute capacity will the system support?
The design targets 10–12 MW of AI compute capacity alongside 15–18 MW turbines and integrated battery storage.
When will the prototype be tested?
A 100 kW prototype is scheduled for testing off the coast of Norway by the end of 2026.
How does the cooling system work?
The ocean acts as an “infinite heat sink.” Heat from the data halls passes through steel tank walls into seawater, minimizing energy use for cooling.
Are there similar projects underway?
Yes. China’s HiCloud has deployed a wind-powered underwater data center off Shanghai, and Japan is piloting offshore floating AI data centers powered by wind and cooled by seawater.
What are the main challenges?
Key challenges include servicing submerged hardware, scaling to gigawatt levels, and ensuring minimal environmental impact.
