Aikido Technologies, an offshore infrastructure provider, has unveiled AO60DC: a first-of-a-kind floating offshore wind platform designed to co-locate AI-grade compute with floating wind generation and integrated battery storage.
The AO60DC is designed to host 10–12MW of AI-grade compute alongside a 15–18MW+ wind turbine and integrated battery energy storage. Designed for farms consisting of 30MW to more than 1GW of IT load, the technology offers a pathway to sovereign, gigawatt-scale AI infrastructure built directly at the source of renewable energy, solving rapid growth in demand for high-density AI infrastructure. Developers find it increasingly difficult to obtain energy, land, and water to build such mega-campuses, especially close to load centers. Offshore, however, energy, cooling and space are abundant and the ocean acts as an infinite heat sink. Around the world, areas pre-designated for floating wind deployments can instead be immediately used for datacenters.
The onboard wind turbine and BESS are designed to power the compute load for the majority of operating hours, with a grid connection used primarily during summer months. Batteries can also be pre-charged ahead of grid stress events, which can significantly shorten effective grid connection timelines for new capacity. The units can be deployed within 200mi (<10ms RTT) of major compute load centers in sovereign waters around the world, allowing many energy-constrained countries to deploy AI infrastructure.
Simplifying integration
“Before we go off-world, we should go offshore,” said Sam Kanner, CEO of Aikido Technologies. “First movers in the O&G industry exploited deepwater resources over 40 years ago and reaped massive benefits. Aikido is well positioned to integrate proven, offshore components with typical data hall construction techniques to build GW-scale AI factories faster, cleaner, cheaper and more efficiently than conventional techniques.”
By combining Aikido’s proprietary wind turbine substructure and the data center enclosure into a single steel unit, the design will reduce both capital and operating costs while simplifying integration. The data halls can be pre-fabricated in a factory and lifted into place during the final integration. The units can be deployed in sovereign waters, leveraging over 50GW of sites around the world, already designated for this type of deployment, streamlining the permitting and grid connection processes.
At the core of the concept is Aikido’s modular “flat-pack” semi-submersible floating platform, which can be assembled up to ten times faster than conventional offshore structures. The approach is intended to support accelerated deployment timelines for AI and energy infrastructure. Semi-submersible offshore platforms are the most universal and proven offshore platform type, with O&G and floating wind operators successfully deploying such platforms at sites around the world for 25+ years.
The platform is expected to reach a power usage effectiveness (PUE) of below 1.08, enabled by a passive primary cooling system that transfers heat through the steel hull into the surrounding seawater. Thermal impact is expected to be limited to a localized area extending only a few meters from the structure. Given the generation of local construction and maintenance jobs, their location far offshore, with minimal visual impact or environmental disturbances, and the potential to provide additional, low-cost power, the datacenters can be net positive for coastal communities and cities.
Deployment later this year
The units can be installed and serviced using vessels already active in offshore wind and deepwater O&G industries, enabling maintenance response times comparable to conventional data centers. The platforms can be manned for days a time to ensure that individual servers have uptime that meet the typical standards in the datacenter industry. The integrated data halls are also engineered to provide the highest level of physical security.
A proof-of-concept unit is currently under development in Norway and is scheduled for deployment later this year. Aikido is a member of the NVIDIA Inception program, and the company has already received early interest from AI inference customers.
The first commercial project is targeted for the UK, with a planned operational date of 2028. A site has already been identified and detailed engineering and commercial discussions are underway.
