The Very Wild, Very Real Plan To Build AI Data Centers In The Ocean - EP 65 Garth Sheldon-Coulson

Pantalassa’s approach centers on a self‑propelled ocean node that turns wave motion into pressurized water flow, driving a single turbine and generator. The device is essentially a hollow, balloon‑like cylinder—about 20 m wide and 80 m long—built with few moving parts, which dramatically lowers maintenance in corrosive saltwater. By eliminating shore‑based infrastructure and using a dam‑style water cycle, the system captures the relentless energy of mid‑ocean waves, a resource that traditional coastal wave converters have struggled to exploit.

Because the nodes are designed for mass production, a single factory can churn out thousands of units that are shipped, self‑deployed, and operated without cables or seabed drilling. This deployment model promises gigawatt‑scale capacity in years rather than decades, outpacing solar’s 50‑year rollout and wind’s site‑specific build times. The technology targets the planet’s high‑energy bands—particularly the Southern Hemisphere’s red‑zone winds—delivering continuous power with far less intermittency than land‑based renewables, which translates into lower levelized cost of electricity.

The immediate commercial focus is floating data centers that house AI chips directly on the node, converting captured wave power into compute energy on site. By locating compute offshore, Pantalassa reduces the need for new land‑based power plants, eases grid congestion, and keeps energy prices stable for terrestrial users. As artificial‑intelligence workloads accelerate toward terawatt‑hour consumption, such ocean‑based power could become a cornerstone of sustainable, scalable compute infrastructure, attracting investors seeking long‑term, low‑carbon energy solutions.