Trump Administration Proposes Transferring Cold War-Era Plutonium Stockpiles to Private Nuclear Energy Startups

The Department of Energy has selected five nuclear startups to enter advanced negotiations for access to approximately 20 metric tons of surplus weapons-grade plutonium from dismantled Cold War-era warheads — enough material to build roughly 2,000 nuclear bombs [1][2]. If finalized, it would be the first time the U.S. government has transferred weapons-grade plutonium to private companies for commercial energy use [3].

The companies — Oklo, Standard Nuclear, SHINE Technologies, Flibe Energy, and Exodys Energy — would convert the surplus material into fuel for advanced nuclear reactors [1]. The plan arrives after decades of failed government-led disposition efforts, including an $8 billion boondoggle at the Savannah River Site in South Carolina, and amid surging electricity demand from the artificial intelligence industry [4][5].

How Much Plutonium, and What Is It Worth?

The United States has declared 54 metric tons of plutonium as surplus to defense needs [6]. Of that total, roughly 20 metric tons would be made available to private companies under the DOE's Surplus Plutonium Utilization Program [2][7]. The remaining surplus is slated for the existing "dilute-and-dispose" strategy, which involves blending the plutonium with inert materials and burying it at the Waste Isolation Pilot Plant in New Mexico [4].

Source: DOE/Arms Control Association

Data as of May 1, 2026CSV

The 20 metric tons in question is weapons-grade material — plutonium with a concentration of the isotope Pu-239 above 93%, making it directly usable in nuclear weapons with minimal processing. According to Oklo CEO Jacob DeWitte, this quantity could generate approximately 2.5 gigawatts of electricity, enough to power roughly 2.5 million American households [5]. In commercial terms, that represents billions of dollars in potential energy revenue over the lifetime of the reactors.

Plutonium itself has no established commodity market price because it has never been commercially traded in the United States. Its value is derived from its energy density: a single kilogram of plutonium, when fissioned in a fast reactor, can produce energy equivalent to roughly 10,000 barrels of oil [8].

The Legal Minefield

The Atomic Energy Act of 1954 provides the foundational legal framework governing nuclear materials in the United States. Under the AEA, the DOE holds authority over "special nuclear material" — a category that includes plutonium, uranium enriched in U-233 or U-235, and any material artificially enriched in these isotopes [9]. The law gives the DOE broad discretion to authorize the transfer, distribution, and use of special nuclear material for peaceful purposes.

However, the U.S. has maintained a de facto policy against commercial plutonium use for five decades, dating to President Gerald Ford's 1976 decision — later reinforced by President Jimmy Carter — to defer commercial reprocessing and plutonium recycling due to proliferation concerns [6]. While this policy was never codified as statutory law, it has guided U.S. nuclear policy through multiple administrations of both parties.

The Trump administration appears to be working within existing DOE authority rather than seeking new legislation. The Surplus Plutonium Utilization Program was launched in August 2025, and the DOE's Reactor Pilot Program, announced July 16, 2025, set an ambitious goal of achieving criticality in at least three advanced reactor concepts outside national laboratories by July 4, 2026 [1][3].

Critics argue this timeline circumvents adequate Congressional oversight. Senator Edward Markey (D-MA), along with Representatives John Garamendi (D-CA) and Don Beyer (D-VA), wrote to President Trump in September 2025 warning that "the transfer of weapons-usable plutonium to private industry would increase the risk of nuclear weapons proliferation, including to rogue states or terrorists" [10].

The Companies and Their Connections

Of the five selected companies, Oklo has attracted the most scrutiny — not for its technology, but for its political connections.

Oklo Inc. is a California-based advanced nuclear technology company with a market valuation exceeding $12 billion despite having zero revenue [5]. Its Aurora fast reactor design uses liquid sodium metal as a coolant rather than water. The company's Pluto reactor project is specifically designed to run on plutonium fuel. In December 2025, Oklo and Los Alamos National Laboratory conducted "fast-spectrum plutonium criticality experiments" at the Nevada National Security Site, demonstrating a key safety feature where the reactor naturally reduces its power output as temperatures rise [3][5].

Oklo's board history is where the political questions concentrate. OpenAI CEO Sam Altman chaired Oklo's board for over a decade before stepping down in 2025 [2][7]. Energy Secretary Chris Wright served on Oklo's board of directors until his Senate confirmation in February 2025, after which he recused himself from decisions involving the company [2][7]. The DOE has stated that Wright "complied with ethics requirements and never owned Oklo stock" [2]. Institutional investors including BlackRock, Vanguard, and Jane Street hold substantial positions in the publicly traded company [3].

Senator Markey has been direct in his criticism, warning that the administration was proceeding because "Oklo stands to benefit financially" rather than out of national interest [2].

Standard Nuclear, SHINE Technologies, Flibe Energy, and Exodys Energy are the other four companies selected for negotiations. Flibe Energy is developing liquid-fueled thorium molten salt reactor concepts. Exodys Energy focuses on modular, proliferation-resistant reprocessing technology. SHINE Technologies, based in Janesville, Wisconsin, has experience with neutron production technology. Less public information is available about Standard Nuclear's specific designs [1][3].

The $8 Billion Lesson: What Happened at Savannah River

The Trump administration's turn to the private sector comes against the backdrop of one of the most expensive failures in DOE history.

In September 2000, the United States and Russia signed the Plutonium Management and Disposition Agreement, with each country pledging to dispose of at least 34 metric tons of weapons-grade plutonium [4]. The U.S. plan centered on a Mixed Oxide (MOX) Fuel Fabrication Facility at the Savannah River Site, which would convert weapons plutonium into mixed-oxide fuel for use in commercial power reactors.

Source: DOE/GAO Reports

Data as of Jan 1, 2024CSV

The project was a catastrophe. Originally estimated at $4.9 billion, the projected cost ballooned to $7.7 billion by 2012, then to over $17 billion by 2016 [4][11]. Construction fell a decade behind schedule. The facility struggled to recruit workers who met nuclear industry qualification standards — a symptom of the broader decay of American nuclear construction expertise after decades without major projects [4]. No commercial utility had signed on as a customer for the MOX fuel [4].

President Obama's fiscal year 2017 budget proposed terminating the MOX project. In 2018, after approximately $8 billion had been spent with the facility only 70% complete, the National Nuclear Security Administration formally killed the contract [11]. The DOE pivoted to the dilute-and-dispose approach, which itself has proceeded slowly.

South Carolina, which had been promised that the plutonium stored at the Savannah River Site would be processed and removed, sued the federal government. The state's frustration captured a broader truth: the U.S. government has spent decades and tens of billions of dollars on plutonium disposition with remarkably little to show for it [4].

Nonproliferation: What the Treaties Say

The Nuclear Non-Proliferation Treaty, signed by 191 states, permits the peaceful use of nuclear energy under International Atomic Energy Agency safeguards [12]. Article IV explicitly recognizes the "inalienable right" of all parties to develop nuclear energy for peaceful purposes. Several countries, most notably Japan, France, and the United Kingdom, routinely separate and use plutonium in civilian reactors [12].

However, the United States has historically held itself to a stricter standard. Since the Ford-Carter policy of the 1970s, the U.S. has discouraged plutonium recycling globally and pointed to its own abstention as evidence of commitment to nonproliferation norms [6]. Transferring weapons-grade plutonium to private companies would represent a sharp break from that posture.

The U.S.-IAEA Voluntary Offer Agreement allows — but does not require — the IAEA to apply safeguards to U.S. civilian nuclear facilities [13]. As a nuclear weapons state under the NPT, the U.S. is not subject to the comprehensive safeguards required of non-weapons states. This means the legal obligation to accept international inspections of any transferred plutonium would depend on whether the receiving facilities are placed on the U.S. eligible facility list — a decision made by the DOE.

There is no public indication that the administration has formally notified or consulted allied governments or treaty partners about the proposed transfers [10]. Edwin Lyman, a physicist at the Union of Concerned Scientists, expressed concern about the broader signal: "If there were adults in the room and I could trust the federal government to impose the right standards, it wouldn't be such a great concern, but it just doesn't seem feasible" [14].

The precedent question is significant. If the United States begins commercial plutonium recycling, it would weaken its diplomatic standing to oppose similar programs in countries like South Korea, Saudi Arabia, or others that have expressed interest in enrichment and reprocessing capabilities.

Jobs, Communities, and the Status Quo

The current custodianship of surplus plutonium supports substantial workforces at federal sites. The Pantex Plant near Amarillo, Texas, employs more than 4,600 people and serves as the primary storage location for plutonium pits from dismantled warheads [15]. The facility has exceeded its permitted storage capacity of 20,000 pits, creating its own set of regulatory and safety concerns [6].

The Savannah River Site in Aiken, South Carolina, employed 13,387 people as of September 2024, with the workforce expected to grow to 18,000 in the coming years as the Savannah River Plutonium Processing Facility ramps up pit production for the NNSA's modernization program [16]. These are well-paying jobs in communities where the federal nuclear complex is the dominant employer.

Relocating plutonium to private facilities would not necessarily eliminate these jobs — Pantex and Savannah River have missions beyond surplus plutonium storage — but it would introduce new risk profiles to whatever communities host private fuel fabrication and reactor operations. The details of where private facilities would be sited, what local consent processes would apply, and how emergency planning zones would be established have not been publicly disclosed.

The Case For: When Government Fails, Try Markets?

Supporters of the transfer make a pragmatic argument rooted in the government's own track record.

The MOX facility consumed $8 billion and produced nothing. The dilute-and-dispose alternative has moved slowly. Meanwhile, 54 metric tons of surplus weapons plutonium sits in aging facilities, costing taxpayers hundreds of millions of dollars annually for security and maintenance [4][6]. The status quo, supporters argue, is itself a proliferation risk — concentrating vast quantities of weapons-grade material in facilities that were designed as temporary storage.

Nick Touran, a nuclear engineer and TerraPower veteran, frames it bluntly: "Plutonium still gets an 'ew' reaction out of the public. But it's undeniably a better reactor fuel. You get more neutrons out when you split a plutonium atom" [5]. In his view, the framing is backwards: "Calling plutonium a bridge fuel to HALEU is inherently misleading. If anything, HALEU should be a bridge fuel to plutonium" [5].

The technical argument is that fast-spectrum reactors — the type Oklo and others are developing — can "burn" plutonium more thoroughly than conventional light-water reactors, converting it into shorter-lived fission products and reducing the volume and longevity of radioactive waste. If successful, this approach would simultaneously generate carbon-free electricity and permanently destroy weapons-usable material, achieving the dual goals of energy production and nonproliferation.

Koroush Shirvan, an MIT nuclear engineering researcher, acknowledged the limits of the available material — 20 metric tons "is not much" and generating more would require reprocessing spent fuel, which brings its own regulatory and proliferation challenges — but concluded: "Is it a morally right thing to do? Yes" [5].

The Case Against: Security, Liability, and What Happens When a Startup Fails

The counterarguments center on security, accountability, and the nature of the companies involved.

Weapons-grade plutonium is among the most dangerous materials on Earth — not because of its radioactivity, which is modest by nuclear standards, but because of its weapons utility. The IAEA considers 8 kilograms sufficient for a nuclear weapon [12]. Twenty metric tons represents approximately 2,500 potential weapons' worth of material [10].

Under existing NRC regulations, private companies handling special nuclear material must maintain physical security plans, personnel reliability programs, and material control and accounting systems [17]. The Price-Anderson Nuclear Industries Indemnity Act provides the liability framework: operators carry $500 million in private insurance per site, with a secondary layer funded by retrospective assessments on all nuclear licensees [18].

But these frameworks were designed for large, established utilities operating conventional reactors — not pre-revenue startups. Oklo's $12 billion market valuation is built on projections, not earnings [5]. If a company holding weapons-grade plutonium went bankrupt, NRC regulations require notification, but the commission's authority to compel continued safety expenditures during bankruptcy proceedings is limited [19]. The practical question of who secures the material during insolvency — and who pays for it — does not have a clear answer in existing law.

The liability gap extends to remediation. If plutonium were lost, stolen, or released at a private facility, the federal government would almost certainly bear the ultimate cleanup cost, just as it has at every contaminated DOE site in the nuclear weapons complex. The Government Accountability Office and the NRC's own analyses have flagged concerns about the financial adequacy of limited liability companies as nuclear licensees [19].

What Comes Next

The DOE's negotiations with the five selected companies are ongoing. Several key decision points remain: the quantity of material each company would receive, the physical security standards that would apply, the role (if any) of IAEA safeguards, and whether Congress will assert oversight authority over the transfers.

The Reactor Pilot Program's July 4, 2026 criticality deadline looms as both a technical and political milestone. If any of the selected companies demonstrates a working plutonium-fueled reactor by that date, it would strengthen the administration's argument that the private sector can accomplish what the government could not. If they do not, it will raise questions about whether the program was designed more to benefit well-connected startups than to solve the plutonium problem.

The broader question is whether the United States can reconcile its half-century commitment to discouraging civilian plutonium use with a new industrial policy that treats weapons material as a commercial asset. The answer will shape not only domestic energy policy but the global nonproliferation regime for decades to come.