Samsung Is Already Building an Industry Around the Idea
The company is not developing the concept entirely on its own.
In 2026, Samsung Heavy Industries announced partnerships with companies including Greek shipping company Capital and Lloyd's Register.
Under that collaboration, Samsung is responsible for technology development and construction, Capital is focused on project sourcing and investment, while Lloyd's Register is involved with rules and regulatory considerations.
Samsung has also partnered with U.S. data center developer M3.
In August 2026, Samsung Heavy Industries and M3 signed an engineering contract covering basic, detailed and production design for floating data centers intended for major U.S. markets, including Texas.
Each proposed facility is expected to provide approximately 50 MW of capacity.
That is significant because it shows the idea is moving from a conceptual rendering toward an actual engineering project.
There is still a major difference between an engineering contract and a fully operational commercial data center.
Construction, financing, permitting, grid connections, equipment procurement and final customer commitments still have to be completed.
But the project is no longer merely theoretical.
The 2028 Target
Samsung Heavy Industries has set the second quarter of 2028 as its target for commercializing its floating data center platform.
The company has said it is pursuing multiple feasible projects with this timeline in mind.
That does not mean the entire data center industry will suddenly move offshore in 2028.
It means Samsung is targeting that period for the commercial launch of its FDC offering.
The distinction is important.
Technology development is rarely a straight line from announcement to mass deployment.
A first facility may demonstrate technical feasibility without immediately creating a large new market.
The economics also have to work.
Customers must be willing to use the infrastructure.
Power must be reliable.
The cooling system must operate safely.
Marine conditions must not significantly reduce equipment life.
Regulators must approve the facilities.
And the total cost must compete with conventional data centers.
The next two years will therefore be important for determining whether floating data centers become a niche technology or a meaningful part of AI infrastructure.
The Ocean Creates New Engineering Problems Too
Moving a data center onto water does not make the engineering challenges disappear.
It creates new ones.
A conventional data center sits on a foundation.
A floating data center moves.
Even if the movement is small, servers and sensitive electrical equipment have to operate reliably despite vibration, inclination and environmental changes.
Samsung itself has acknowledged that offshore environments can expose AI servers to vibration, inclination, high-salinity air and rapid changes in humidity.
Salt is another major issue.
Seawater is highly corrosive.
Salt-laden air can damage metal structures and electronic equipment if it is not properly controlled.
That means the facility needs sophisticated protection against humidity and salt intrusion.
Samsung says it is developing technologies to address these issues, including systems for offshore positioning and protection against salt and humidity. Its partnership with Supermicro also includes operational verification of AI server infrastructure in marine and river environments.
These are not minor technical details.
A data center is expected to operate continuously.
Replacing damaged servers offshore is considerably more complicated than replacing a server in a conventional building.
Maintenance Could Be One of the Biggest Questions
This may become one of the most important tests for the concept.
A land-based data center is designed so engineers can walk into the building whenever necessary.
They can replace equipment.
Inspect cables.
Repair cooling systems.
Install new servers.
A floating facility can still be accessible, especially if it is near shore.
But farther offshore, maintenance becomes more complicated.
The ocean introduces weather conditions, waves and transportation challenges.
A serious equipment failure could require specialized marine crews or temporary shutdowns.
This is one reason Samsung's nearshore approach makes sense as an initial step.
It provides many of the advantages of water-based infrastructure while keeping the facility relatively close to conventional support systems.
Over time, more autonomous monitoring, robotics and remote maintenance could potentially make farther-offshore facilities practical.
But that technology would have to become reliable enough for infrastructure that may be carrying extremely valuable AI workloads.
There Is Already a Precedent for Computing Underwater
Samsung's concept is not the first attempt to move computing infrastructure into the ocean.