Others buy what the Navy only tries
In May 2023 the Royal Navy and Imperial College London tested a quantum accelerometer in a pod aboard XV Patrick Blackett. The sensor uses cold atoms to measure a change in speed, and it is the kind of instrument that might one day let a ship find its way with no satellite fix at all. The pod was one of a set of swap-in bays that QinetiQ designed, so devices could go in and come out again. The ship’s captain, Commander Michael Hutchinson, said the technology was “still in its very early stages”. He was right about more than the sensor.
Three and a half years later, Britain’s defence quantum effort still looks like that pod. It is a well-made bay that other people’s devices pass through. At least twelve field tests of quantum hardware with the forces, Dstl (the ministry’s research laboratory) or a defence firm have been published since 2023, and the ministry has published no order to buy any of it in numbers. Britain has become very good at trying quantum. The established firms that lead its navigation and timing research, meanwhile, have American parents.
The public registers make the point bluntly. Search Contracts Finder, the government’s list of public contracts, for the word quantum and for the sensors in a brain scanner, and it holds ten Ministry of Defence and Dstl awards since 2022. Together they are worth about £4.3 million. The largest, £2.8 million to Cerca Magnetics in November 2025, pays for one mobile brain scanner that Defence Medical Services will use to study the effect of blasts on troops. The eight awards that name quantum outright come to £948,000, and the last of them is dated November 2023.
The ministry has a fair reply, and it deserves a full hearing. Defence and security contracts can be exempt from publication, and work bought through a long deal with a prime contractor such as QinetiQ would carry the prime’s name and not the word quantum. The ministry says it has put more than £28 million into quantum research through Dstl, though it gives no period for the sum. A minister told Parliament in September 2026 that the department holds no single figure for quantum investment. On this reading the registers are a floor. The real effort is larger, and quieter.
Caution suits a fragile technology. Commander Hutchinson’s warning was fair, since a sensor that behaves in a laboratory still has to prove itself on a rolling deck. The National Cyber Security Centre, part of GCHQ, has been just as careful about quantum key distribution, a way of sharing secret keys over a quantum link. In a white paper of March 2020 it said it does not endorse the method for any government or military use, and that position still stands. The method does not prove who is on the other end of the line, and it rests on hardware that is hard to vouch for. A buyer that tests before it signs is doing its job.
Yet a test is supposed to end in a decision, and the record shows very few. In January 2025 Dstl announced an optical atomic clock that the ministry called the first of its kind built in Britain, adding that it could be on operations within five years. It would lose less than a second over billions of years, the ministry said. That is the only precision figure any of the dozen published trials has offered, and it is the ministry’s claim rather than a drift measured against a reference clock. In February 2026 Dstl went back to the field. It gave no figure for drift at all.
The pattern repeats down the list. Aquark Technologies of Hampshire took its AQlock clock back to sea on HMS Pursuer in June 2025 and said it gave stable outputs, but it published no drift number. In October 2025 Infleqtion’s Tiqker clock ran aboard Excalibur, the Navy’s test submarine with no crew. Commander Matthew Steele of the Navy’s new-technology office called it a first step towards quantum clocks on long trips under the sea. Infleqtion gave no measured figure either. None of the dozen trials published since 2023 has given a measured accuracy or drift figure.
The most telling trial is one that never reported. Infleqtion’s British arm led Q-NAV, a £3.5 million research contract run with QinetiQ and the Navy and paid for by Innovate UK, which is part of the state funder UK Research and Innovation. The aim was to fuse classical and quantum motion sensors into one system that holds a course when satellite signals are lost. Infleqtion said in August 2024 that the first phase was done. A sea trial on the Patrick Blackett was to follow in early 2025. The project’s public record now lists it as closed with no outcome posted. No result from the trial has been published.
Q-NAV was not Ministry of Defence money. Yet that one Innovate UK contract is worth about four-fifths of all ten ministry and Dstl awards on the register combined. The comparison flatters Q-NAV, since some ministry spending never reaches the register at all, but it still shows where the visible money in British defence quantum comes from. It is research money, and it buys research. The ministry has bought kit only one machine at a time, ORCA’s PT-1 in 2022 and Cerca’s scanner in 2025, and both were single machines bought for study. Even ProQure, a computing competition in a March 2026 package, pays first to build and test machines. Its page says it does not buy any solution.
Computing tells the same story with a longer memory. The ministry’s purchase of ORCA Computing’s PT-1, a machine that computes with particles of light, was announced in June 2022, and ORCA and the BBC called it the government’s first quantum computer. Dstl’s own review says it bought one machine, to be housed at the National Quantum Computing Centre, and no price has been published. In 2020 Dstl had said a first quantum computing capability could be in service by 2025 if tests bore it out. None has been announced. In February 2026 Oxford Quantum Circuits, or OQC, said QinetiQ had run its own algorithm on Toshiko, an OQC machine, to find the weak points in a battlefield radio network. OQC’s chief executive, Gerald Mullally, called it “a tangible example of quantum computing’s power to deliver real operational value”. OQC’s own case study was more modest. It said the networks tested were still small enough to solve on a normal computer.
Then ask who owns the firms whose devices go into the pod. In 2023 Innovate UK started six navigation and timing contracts worth more than £3 million each. Three American-owned firms led four of the six. Infleqtion, whose parent is American, led two, and TMD Technologies of Hayes, bought by CPI of the US in 2021, led another, a navigation system called HARLEQUIN. In February 2024 Britain let TransDigm buy CPI’s British units, but only if atomic clock research and making stayed in the country. Microchip Technology Caldicot led a quantum gyroscope project, and its parent is Microchip of Arizona. The habit is older than that. From 2018 to 2021 Teledyne UK of Chelmsford, owned by Teledyne Technologies of the US, led a £4.45 million project to build a small atomic clock.
None of these firms did anything wrong, and all four are run from Britain. Infleqtion’s British arm has been in Oxford since 2014, and in 2026 it began a centre there to add research and factory space. Its parent is American, though, and it sells to the Pentagon. It took a $2 million US Army award in December 2025 and a US Navy deal in April 2026, both for machine-learning software rather than quantum hardware, and both from the American forces. IonQ said in September 2025 that it had completed its purchase of Oxford Ionics. Quantum Motion, another British firm, is in the second stage of a benchmarking scheme run by DARPA, the American defence research agency.
This is the real cost of testing without buying. A British trial gives a firm a reference, and the firm can carry that reference to any buyer, at home or abroad, who wants proof that the kit has left the laboratory. Among the makers in this story, the established navigation and timing firms have American parents. The British-owned makers are younger firms such as Aquark, Delta g of Birmingham and Zero Point Motion of Bristol. The state has its own venture fund, the National Security Strategic Investment Fund, whose published list holds seven quantum firms, two of them foreign. A venture stake buys a seat at the table. It does not keep a firm alive the way a production order does.
The young British firms show what the alternative might look like. In June 2026 Aquark placed two of its AQlock clocks at separate sites, and they kept two radars, a Saab Giraffe 1X among them, on one shared air picture with no satellite timing. From cold, they reached a usable timing signal in under 30 minutes. Aquark called it the first known use of a pair of British-built cold-atom systems on a real defence problem. Its seed round in September 2024 was led by the NATO Innovation Fund, with MBDA among those investing alongside. In May 2026 the ministry gave Aquark a Commercial X contract through its fast-track buying route. Neither side has said what it is worth.
The need is not in doubt, because satellite signals can be jammed or faked, and a 2023 study for the government put the cost to the country of a 24-hour loss of them at £1.42 billion. Britain’s own national mission says quantum navigation, clocks included, is to be on aircraft by 2030. A submarine cannot lean on satellite timing, which is why the Excalibur trial mattered. The ministry knows all this, which makes the gap between its trials and its orders harder to explain.
Policy has started to promise something different. From 2026, says the Defence Industrial Strategy of September 2025, at least 10 per cent of what the ministry spends on equipment will go on new kinds of technology, with quantum named among them. The Defence Investment Plan of June 2026 gives UK Defence Innovation, the ministry’s innovation arm, £1.6 billion over four years. The plan says that includes “embedding and expanding AI and quantum across the Armed Forces”. Neither paper puts a sum on quantum. A promise to buy new kit, with no figure for the kit in question, is a pod with no device in it.
For a supplier the lesson is uncomfortable. A Navy trial is a reference, and a firm counting on ministry revenue should plan on Dstl research work and Innovate UK contracts until something changes. An investor reading defence as the route to market has four years of trials and no published order to go on. The next tests of that are already dated. Dstl plans another field trial in 2027, the national navigation hub runs to 2029 and the aircraft target falls in 2030.
One document would overturn this argument. It would be a ministry contract to buy quantum clocks, sensors or computers in numbers, published on a public register beside the PT-1 and the ten small awards that make up today’s record. None has been published, and until one is, Britain will keep building test bays. Others will keep buying the firms that fill them.
See today’s quantum computing news on Quantum Zeitgeist for the latest breakthroughs in qubits, hardware, algorithms, and industry deals.
