Nuclear Power Just Went Mobile. The US Navy Got There First.

Week 31 | July 2026

Every disaster response plan makes the same bet: that fuel and generators are already sitting somewhere near where the disaster is about to happen. They rarely are. Puerto Rico waited months for grid power after Maria. Fiji's outer islands wait just as long after every cyclone season, because pre-positioning diesel across scattered Pacific islands is not something any small nation's budget survives. What the disaster logistics world actually wants doesn't exist yet: a power source that can travel to the emergency instead of waiting to be found by it.

Just before midnight on 30 June in Idaho, that idea got one component closer to reality. Deployable Energy's Unity microreactor, a one-megawatt unit built to fit inside a standard shipping container, reached criticality at Idaho National Laboratory, the third DOE-authorised advanced reactor to do so under the compressed 2026 deadline. Four days later, on the Fourth of July, a second company, Aalo Atomics, put its own test reactor critical at the same site, the fourth advanced design to clear that bar under the same deadline. Two reactors, four days apart, one laboratory. Call this a regulatory sprint that is working.

Unity's first real maritime customer is a ship. Seatransport, an Australian shipbuilder run by Stuart Ballantyne, has partnered with Deployable Energy and Lloyd's Register (LR) to fit a 73-metre amphibious landing vessel with two to five of these microreactors, targeting eight to ten years at sea without refuelling. LR issued Approval in Principle for the design on 26 February this year, the naval architecture equivalent of a green light to keep spending on engineering, rather than a permit to build. Jez Sims, LR's nuclear technical authority, was careful to say this approval “is not a guarantee.” The first live reactor test on an actual hull isn't scheduled until November or December next year. Everything about this program is real. Very little of it is finished.

Navy Lookout's review of the Seatransport design puts the crew at 18 to 21 people, and every public reference to this ship, from Ballantyne's own writing to LR's press materials, assumes a bridge with humans standing on it. Nobody in this program has proposed removing the crew. Nobody has proposed a fleet of these vessels either, whatever might be implied by looser trade coverage. The headline number that does get quoted, 100,000 units by 2040, belongs to Unity as a reactor product being sold to mining companies and data centre operators. That's a different count entirely and a much bigger universe than ships alone.

Which raises a question worth asking directly, even if nobody building these systems has asked it yet: is there scope for a nuclear-powered, uncrewed vessel to act as a disaster relief ship in its own right? Two enabling pieces already exist, just not together. LR runs a six-level autonomy framework, AL1 through AL6, for uncrewed and remotely operated vessels. Nobody has applied it to a nuclear hull. Nobody at LR, Seatransport, or Deployable Energy has said they should. 

The self-deploying reactor that sails itself to a blackout and plugs in with no crew aboard is OTI's own extrapolation of where two separate engineering tracks, nuclear propulsion and vessel autonomy, might eventually converge. Nobody is building or regulating it yet. The disaster relief problem it would solve is already real. Whether this is the solution anyone actually wants is a separate question, and so far, nobody's asked for it.

BWXT shipped 40,000 TRISO fuel compacts to Idaho's Transient Reactor Test Facility last November for Project Pele, the Army's transportable microreactor program, and the Army's follow-on Janus Program is already picking candidate bases. Dr Jeff Waksman, the Army's principal deputy assistant secretary, called it delivery of “real nuclear microreactor fuel,” not another promise on paper. On 14 May this year, acting Navy Secretary Hung Cao told Congress that Norfolk Naval Station will draw shore power from USS Gerald R. Ford's reactors this summer, a live trial of the same ship-to-grid transfer the disaster relief vessel would eventually need. As at time of writing, no results of that trial have been published.

Some may ask about weapons risk, so here's the direct answer: Unity runs on uranium enriched to 4.95 percent, standard low-enriched fuel, below the enrichment levels that drive most of the current proliferation debate about newer HALEU designs like Pele's. That's genuinely reassuring, and it deserves to be said plainly instead of buried in a footnote. What isn't settled is the legal status of a reactor with no permanent crew (were one to be proposed). Would it be classed as a vessel, an installation, or something maritime law hasn't defined yet? And which state's courts have jurisdiction if something goes wrong in international waters? The IMO is rewriting its 40-year-old nuclear merchant ship code right now, with industry summaries pointing to a target adoption around 2030. That gap between the engineering timeline and the legal one tells you most of what you need to know.

One more thing worth naming. Fiji's government has floated hosting a regional disaster response vessel under its Ocean of Peace initiative, but no humanitarian agency has actually asked for this technology. UNHCR's disaster energy work is solar and diesel hybrids. OCHA, IFRC, and the ICRC don't mention nuclear power in their relief planning, and when the ICRC does use the word nuclear, it means weapons, not generators. The demand signal here is coming entirely from shipbuilders and reactor vendors. That doesn't make the idea wrong. It means nobody has said yes to it yet.

Our institutional gap thesis holds again. Containerised reactors, autonomous hulls, and disaster power logistics are closing from three directions that haven't yet been asked to talk to each other. The vessel that exists is crewed and years from its first reactor test. The autonomy exists on an entirely different class of ship. The law governing the combination doesn't exist at all. Whoever writes that law first gets to decide who builds the thing this brief just spent a thousand words describing as not yet real.

A US Navy Rear Admiral told Congress the Navy wants half its surface fleet unmanned within two decades. That's over 100 medium unmanned surface vessels, plus an unknown number of smaller ones, operating alongside crewed ships. Everyone covered the announcement. Nobody's covered what that fleet actually looks like, architecturally and industrially, once you build it. We take that up next week. 

Since you have been, thanks for reading.

Cheers,

Mick

Ocean Tech Intelligence provides informational analysis only. Nothing in this publication constitutes financial, investment, legal, or strategic advice. Readers act on this content at their own risk. For full details see our Disclaimer.