Trump orders U.S. to build a quantum computer by 2028 — and to defend against one

The White House issued twin executive orders on 22 June 2026 directing the United States government to accelerate development of a large-scale quantum computer capable of breaking current public-key encryption, while simultaneously mandating a faster migration of federal systems to post-quantum cryptographic standards. The orders, signed at 21:12 UTC, set a target of 2028 for a working "powerful" U.S. quantum machine and frame the effort in the language of national security and industrial competition [1][3].

The orders amount to a public bet that the federal government can compress a research-and-engineering programme of historic difficulty into roughly two and a half years, while also preparing every agency whose systems rely on RSA, elliptic-curve and Diffie–Hellman key exchange to operate in a world where those primitives no longer hold. The political framing — "American quantum dominance" — and the technological ambition are both stated without qualification.

What the orders actually direct

The build-side order instructs the federal research and procurement apparatus to align spending, contracting authority and laboratory roadmaps behind a 2028 deliverable. According to a White House tease circulated earlier in the day at 18:34 UTC, the order is being sold as a guarantee of "American quantum dominance" — a phrase whose commercial and strategic assumptions are doing the heavy lifting [4].

The defence-side order addresses the uncomfortable second half of the same problem. A machine capable of running Shor's algorithm at scale would render most of today's internet- and document-encryption obsolete. Migrating federal systems to post-quantum cryptography — lattice-based, hash-based and code-based schemes that have been standardised by the U.S. National Institute of Standards and Technology over the past two years — is a multi-year programme that, on the government's own previous timelines, was not expected to complete before the mid-2030s. The new order accelerates that clock [1].

The pairing is the point. A build mandate without a migration mandate would leave U.S. federal communications exposed to the first adversary to reach the threshold; a migration mandate without a build mandate would amount to unilateral disarmament in a race the government has already said it intends to win.

The state of the actual hardware

Quantum computing has spent two decades moving between "breakthrough" announcements and the humbler realities of qubit counts, error rates and interconnect bottlenecks. The companies best positioned to deliver a useful large-scale machine — IBM, Google, Microsoft, IonQ, Rigetti, Quantinuum and a handful of Chinese groups including the University of Science and Technology of China — measure progress in a mixture of physical qubits, logical qubits and a less glamorous figure: the error-corrected gate operations per second that determine whether a circuit can do useful work before decoherence.

A useful cryptanalytic machine — one capable of running Shor's algorithm against RSA-2048 in hours rather than the lifetime of the universe — requires something on the order of millions of physical qubits once error correction is included, not the hundreds to low thousands that leading systems currently operate with. No public roadmap from any of the major Western labs has committed to that threshold inside two and a half years. Several have committed to milestones that, if hit, would constitute important intermediate steps toward it.

The gap between the political target and the engineering target is not a reason to dismiss the orders. It is the reason to read them carefully.

The China variable

The orders cannot be read without the parallel track in Beijing. China has treated quantum information as a strategic priority for more than a decade, funding national laboratories, satellite-based quantum key distribution experiments and a domestic supply chain for dilution refrigerators and control electronics. Western analysts have repeatedly concluded that Chinese progress in quantum communication and in superconducting qubits is competitive with — and in some narrow areas ahead of — the United States.

That competitive pressure is the structural backdrop to a 2028 deadline. Two and a half years is, roughly, one U.S. electoral cycle; it is also one round of major procurement decisions. The order effectively tells agencies which side of that cycle they are allowed to be on, and which vendors are eligible to receive the spending that follows.

It is worth registering what this framing leaves out. A U.S. lead in a single cryptographic primitive does not, by itself, decide the geopolitical balance in quantum computing. China publishes openly on its post-quantum standardisation work through the Chinese Association for Cryptologic Research; its industry has shipped commercial quantum communication networks. Whether the U.S. order accelerates a real American lead or simply accelerates spending is a question the orders themselves cannot answer.

What it costs and who pays

The orders are spending directives as much as they are technology directives. Federal quantum-adjacent procurement already runs through the Department of Energy national laboratories, the National Science Foundation, the National Institute of Standards and Technology, the Defense Advanced Research Projects Agency, and the Intelligence Advanced Research Projects Activity, each with its own contracting preferences and its own incumbent vendors. A 2028 deadline is, in practice, a directive to consolidate those pipelines behind whichever companies can deliver credible interim milestones.

The political economy of that consolidation is straightforward: IBM, Google, Microsoft and a small group of well-capitalised pure-plays will collect the largest contracts. Smaller academic groups will find themselves either absorbed into the prime contracts or competing for the residual funding that follows.

A separate consideration sits alongside the build programme. The migration of federal systems to post-quantum cryptography is a defensive mandate whose cost is borne mostly by agency chief information officers, by the contractors who maintain their systems, and by the supply chain of hardware security modules and certificate authorities that underpin current public-key infrastructure. Those costs are diffuse, harder to rally behind politically, and easier to defer. The order names them but does not, on the available reporting, attach a comparable deadline.

The president's own portfolio

It is hard to ignore, in this context, that on 22 June 2026 — the same day as the order — the president publicly remarked that he had previously owned shares in IBM and "brilliantly sold" them when he took office, conceding in the same breath that the sale "was not a good move" [2]. The comment, made on the same day his administration was issuing orders to direct federal quantum spending, illustrates a recurring pattern of the administration's industrial-policy messaging: the political class is happy to pick winners in public markets in language that suggests a personal financial angle to the picking.

Whether IBM is one of the beneficiaries of the new quantum procurement push is a question the orders themselves do not settle. That IBM is named in the president's own remarks on the day of the signing is the kind of detail disclosure regimes in other jurisdictions are designed to surface.

What remains uncertain

Three things are not pinned down by the orders themselves or by the public reporting around them.

First, the precise hardware specification. "Powerful" is doing work that "fault-tolerant" and "cryptographically relevant" would otherwise do. The 2028 target is a political date; the engineering date — the moment at which a logical qubit count is high enough and an error rate is low enough to credibly threaten RSA-2048 — is set by physics, not by executive order.

Second, the migration programme. A deadline of 2028 for the build side implies, by symmetry, an urgency on the defensive side that the available reporting does not yet pin down. Migrating federal systems to post-quantum cryptography is a multi-year programme with hundreds of federal systems, thousands of contractors, and a long tail of legacy code that was not designed to swap its primitives. Whether the defensive order attaches a comparable deadline to that work, or whether it stops at the level of policy direction, is the kind of detail that determines whether the orders are operationally serious.

Third, the international coordination. The United States has, through NIST, taken a leading role in standardising post-quantum cryptographic algorithms. Whether the new orders extend that coordination — or whether they treat post-quantum migration as a unilateral American problem — will shape whether allied governments adopt the new American timelines or set their own. The orders, on the public record, are silent on this.

Stakes

If the orders hold to their 2028 target, the United States will have set the most aggressive publicly stated quantum-computing deadline in the world, will have consolidated a meaningful share of federal quantum spending behind a small number of vendors, and will have committed to migrating its own cryptographic infrastructure on a timeline that no peer government has yet matched. The asymmetric risk is that the public deadline becomes the metric by which the programme is judged, regardless of whether the underlying hardware and migration programmes actually meet the engineering bar.

The political incentive to declare success in 2028 is obvious. The engineering incentive to declare success honestly is less obvious. The orders have placed those two incentives on a collision course.

A counter-reading worth weighing

The most credible counter-reading is that the orders are, in essence, a procurement commitment dressed in national-security language, and that the strategic payload is the redirection of federal research spending toward a small set of pre-selected vendors rather than the literal delivery of a cryptographically relevant machine in two and a half years. On this reading, the 2028 date is the hook that justifies the spending; the real product is the industrial-policy alignment.

The case against this reading is that the migration order is hard to fake. A federal cryptographic migration either completes or it does not, and the systems that depend on it either work or they do not. If the migration order carries real deadlines and real accountability, the build-side 2028 target has a defensible interpretation even if the literal machine slips: the United States will be measurably more quantum-resilient in 2028 than it is in 2026.

Which reading prevails is the question the next twenty-four months will answer.

Desk note

The wire reporting on these orders has so far led with the political framing — "American quantum dominance," a 2028 deadline, a presidential signing ceremony — rather than the engineering substance: the qubit counts, the migration timelines, the procurement machinery. Monexus is tracking the order text as it becomes available and will report on the technical specifications as they are disclosed.

Sources are listed below in the article record. This piece was prepared from public reporting and White House statements; the engineering specifications referenced are drawn from the public roadmaps of the named companies and from NIST's post-quantum standardisation programme.

Wire provenance

This editorial synthesis draws on the following public wire/social posts:

• https://x.com/unusual_whales/status/quantum-ibm-2026-06-22
• https://x.com/polymarket/status/quantum-eo-2028-2026-06-22
• https://x.com/polymarket/status/quantum-dominance-tease-2026-06-22
• https://x.com/unusual_whales/status/hormuz-tankers-2026-06-22