Trump's Plutonium Gambit Reveals the Hollow Core of American Leverage

The administration wants to convert weapons-grade plutonium into reactor fuel — a policy framed as energy security that reads, more honestly, as an attempt to monetise the empire's Cold War arsenal while the getting is good.

By Moemedi Michael Poncanaamericas4-minute read27 May 2026☆ Save ↗ Share ⎙ Print

The Trump administration is moving forward with a plan to convert weapons-grade plutonium from retired nuclear warheads into reactor fuel — a policy the New York Times reported on 26 May 2026 as an attempt to put decades of surplus fissile material to commercial use. The Department of Energy wants nuclear startups to take material that has sat in storage for years, potentially displacing Russian and other foreign supply chains for civilian nuclear fuel. The framing is clean: solve a weapons-disposal problem, build a domestic industry, stick it to Moscow. It is also, characteristically, incomplete.

A former senior US official speaking to Middle East Eye on 27 May put the transactional logic plainly: Trump's pitch rests on a mistaken impression that the countries he ensnared into a war owe him a favour. The comment was made in a regional context, but the principle travels. Washington has spent three years organising coalitions around a Ukraine conflict that has reshaped European energy architecture, defence spending, and industrial competition simultaneously. The administration now wants those arrangements to yield dividends — LNG contracts, weapons purchases, rare earth access, diplomatic alignment on China. The plutonium plan is, at one level, the same ask dressed in technical language: take our inventory, fund our contractors, trust our supply chain.

The policy is not without merit on its own terms. The US government has sat on dozens of tons of weapons-grade plutonium for decades, a storage liability with proliferation risk attached. Offloading it into reactor fuel through advanced designs — potentially mixed-oxide or fast-reactor configurations — solves a genuine national security problem. The nuclear startups being recruited for the task are, by definition, younger and less risk-averse than the established utilities that have historically dominated civilian nuclear procurement. If the technology works, the US could displace Russian fuel supply to allied电网 markets in Eastern Europe and Southeast Asia, where Rosatom and TVEL have maintained share partly because Western suppliers abandoned the segment during the post-Cold War drawdown.

But the merit argument is not the only argument being made, and it is not the one driving the timing. Trump has consistently treated alliances as client relationships: you signed up, you got protection, now you pay. That is not how allied governments — in Berlin, Warsaw, Tokyo, Riyadh — actually experience the arrangement. They signed up for collective security, not for a subscription to American industrial policy. When the State Department or National Security Council presents the plutonium conversion programme as a way to "strengthen energy security for partners," the partners hear a commercial offer wrapped in strategic language. That gap — between what Washington thinks it is offering and what recipients think they are receiving — is the structural problem the administration's entire foreign policy posture keeps running into.

On AI, meanwhile, the Polymarket market implies roughly a one-in-eleven chance that Trump orders a federal review of new AI model releases by 31 May 2026. That is not a prediction; it is a probability estimate reflecting current uncertainty about the administration's direction on compute governance. But the market itself is worth examining. It exists because observers cannot tell, from public statements and institutional signals alone, whether the administration has a coherent position on frontier AI risk or is using safety language as a cover for industrial targeting of Chinese labs. The ambiguity is not accidental. A review mechanism that covered Chinese models but exempted American ones would look like export control extension; one that applied symmetrically would face opposition from OpenAI, Anthropic, and their investors. The administration has not resolved which version it prefers, and Polymarket participants are pricing that indecision.

The plutonium story and the AI story are not unconnected. Both involve the administration using regulatory and procurement levers to reshape global supply chains in ways that benefit American incumbents. Both involve framing decisions as technical or safety-motivated when their primary effect is to consolidate leverage over allied dependencies. Both reflect a foreign policy that has abandoned the multilateral architecture — the IAEA, the NSG, the Wassenaar Arrangement — in favour of bilateral dealmaking that makes American partners bear the adjustment costs.

The countries being asked to accept these arrangements are not wrong to notice the pattern. They were told they were joining a rules-based system; they are being offered a vending machine. The plutonium conversion programme may well produce cleaner energy, stronger allies, and a smaller weapons stockpile. It might also produce American nuclear firms with guaranteed demand and partner governments with expensive contracts they did not need to sign. The administration will tell you the two outcomes are the same thing. Its partners should be slower to agree.

Monexus published a straight news version of the plutonium programme on 26 May, citing the New York Times as primary source. This opinion piece examines the policy logic and its implications for allied relationships.