In October 2025 an uncrewed submarine called XV Excalibur made a series of dives carrying a quantum optical clock, and kept accurate time throughout without a single GPS signal reaching it. The clock was Infleqtion’s Tiqker; MSubs built the vessel; the Royal Navy ran the trial. Rather than taking a time signal from satellites overhead, the clock reads the behaviour of individual atoms, a reference that needs nothing from space and cannot be jammed from the ground. It was the first time such a device had been operated underwater at sea. The trial was delivered six months ahead of schedule, and Excalibur was handed to the Royal Navy on 11 December 2025 to begin two years of sea trials.
A different machine elsewhere in the UK runs on a version of the same idea. Cerca Magnetics, a University of Nottingham spinout formed in July 2020, installed a wearable brain scanner at Young Epilepsy’s Health and Research Centre in 2021: the world’s first clinical wearable MEG system, built for children. Its sensors are optically pumped magnetometers, arrays that also use atoms rather than a satellite or a moving part as their reference. Where the submarine’s clock tells the vessel what time it is, the scanner’s array tells clinicians where a seizure begins.
Same underlying physics. Two very different speeds.
Why the Navy moved
The urgency behind the submarine trial is not hypothetical. Lithuania’s Communications Regulatory Authority recorded 1,022 satellite navigation interference incidents affecting aircraft in June 2025, against 46 in June 2024, and identified more than ten transmission sites inside Russia’s Kaliningrad exclave. A working paper filed jointly by Estonia, Finland, Latvia, Lithuania, Poland and Sweden to the ICAO assembly on 10 September 2025 put the share of flights affected across their airspace at around 27.4 per cent in April 2025, up from about 7.5 per cent a year earlier.
It has since got worse. Poland’s National Institute of Telecommunications logged full-scale disruption on 63 per cent of days across the first nineteen days of August 2026, and on roughly seventy per cent of days through May and June, attributing it to electronic warfare systems operating in the Baltic region. Europe’s institutional answer has been steady rather than quick. Galileo’s OSNMA authentication service went live on 24 July 2025, and EASA and EUROCONTROL published a joint action plan on interference on 26 March 2026.
No accident has yet been attributed to the jamming, and attribution itself turns out to be contested. On 24 September 2025 Spain’s defence ministry confirmed that Margarita Robles’s aircraft suffered GPS interference near Kaliningrad on the way to Lithuania. Three weeks earlier, Ursula von der Leyen’s flight into Plovdiv was reported across Europe as jammed, on information the European Commission said had come from Bulgarian authorities. Flightradar24 then stated the transponder had shown good GPS quality from take-off to landing; Bulgaria’s deputy prime minister denied passing on any such information; the aircraft had in fact landed on a standard ground-based approach. One case stands. The other dissolved on inspection.
That is the awkward part of the evidence. Aircraft fall back on inertial and ground-based navigation and almost always land uneventfully, so a rising threat produces very few incidents anyone can point at afterwards. The money moved anyway. Infleqtion opened an Oxford innovation centre and manufacturing hub on 28 May 2026, expanding UK production of the clock. Interference is above all an aviation problem, and a submarine is simply where the atomic clock happened to be proven first.
What the scanner is still waiting for
The medical side started earlier and has moved slower. Cerca’s system has been installed and generating research data since 2021. Young Epilepsy, University College London and Nottingham, who developed it together, describe it as more sensitive and more comfortable than the conventional MEG and EEG equipment it would replace, and it lets a child move during a scan without sedation. The Epilepsy Research Institute puts the potential saving at around £2 million a year per site, on the assumption of two additional epilepsy surgeries a week.
None of that has reached routine care. Cerca states plainly that its OPM-MEG system is a research system, with no medical or regulatory approval in any jurisdiction. In April 2026 the company raised £3.8 million in Series A funding, earmarked for clinical approval and manufacturing scale-up. The Epilepsy Research Institute has been running an Innovate UK project with Cerca and Nottingham since September 2025 to map an NHS adoption route, and there is still no clinical pathway at the end of it.
The two devices are not interchangeable, and it would be dishonest to pretend otherwise. An optical clock and a magnetometer array do different jobs, and proving one underwater tells you nothing about approving the other for use on a child’s brain. What they share is a starting point. Both stopped depending on a fragile external reference, a satellite or a moving part, and started reading atoms instead. Both worked. What separated them afterwards was not the physics.
Defence had one buyer who could say yes. The Royal Navy identified a risk that was already escalating and already costed, held the budget, and needed nobody’s permission to run the trial. That is why Excalibur finished six months early. Nothing in the chain required a second signature.
The NHS has no equivalent. No single person can authorise a scanner into paediatric neurology, because the authority is distributed across regulatory approval, clinical evidence, commissioning and trust-level budgets, and each of those can only say no or not yet. That distribution is not a fault in the system. It is the design, and it exists because a device that misreads a child’s brain causes harm that a slow procurement never does.
But the design has a cost, the cost is time, and someone pays it.
My Opinion
The asymmetry here is not about whether the technology works. It is about who holds the authority to act once it does. Defence concentrates that authority in one place; health distributes it deliberately, for reasons I would not want reversed. What follows is that a medical device must prove itself to every gate in turn, while a single defence buyer facing a live threat can decide the proving is finished. I am not arguing for fewer gates. I am arguing that we should be honest that they are a choice with a price, and that the price is invisible in a way the Baltic jamming figures are not.
A submarine can now find its way without a signal from space, and the state that wanted it had every mechanism it needed to make that true inside a year. The scanner that could show a clinician exactly where a child’s seizure starts has been sitting in a research suite since 2021, still without the approval that would let it be used in that child’s care.
You’re reading The Next Evolution by Neil Catton, articles that explore the human world and the intersection of technology, they try and ask difficult questions - not to scare - but to inform. If someone forwarded this to you, you can subscribe free at neilcatton.substack.com.
Neil Catton is the author of The Next Evolution, The Cognitive Crucible and The Shadow System - available on Amazon, and writes at the intersection of technology, ethics, and human purpose.