Subscribe
Share
Share
Embed
The UK FCA launches a regulator-first quantum group to test quantum detection of money laundering and fraud and molecules and junctions. A morning brief on what changed and what it means.
Note: The error-correction story in this episode is Riverlane's research, as reported by Quantum Computing Report. An editing step removed the sentence that named them.
Chapters
00:00 Today's headlines
00:49 What quantum advantage actually means
02:36 Error correction with augmented priors
03:25 Perovskite spin lifetime stretched
04:28 Laser-cooled CaF molecules entangled
05:45 Re-engineering Josephson junctions
07:02 Posted on X: government to buy Moon gas
07:48 Quantum Magic School exhibition in Busan
08:24 UK FCA Quantum Experimentation Group
10:58 Hype Check of the FCA-led group
Source links
Quickly Quantum is an AI-voiced podcast, built and run by a real person. Nothing in this episode is financial advice.
Follow Quickly Quantum: subscribe on YouTube · Apple Podcasts · Spotify
More from Brian Lampert: Concrete Compute, the daily AI infrastructure briefing, and Space Stakes, the business of the new space race. Transcripts and every episode: quickly-quantum.kngoworld.chatgpt.site.
The daily quantum computing briefing you don't need a physics degree to follow. Every day, Quickly Quantum cuts through the hype to bring you the breakthroughs, funding rounds, policy moves, and research that actually matter, with plain-English analysis a smart non-physicist can follow. Hosted by Brian Lampert. AI-voiced, human-built, always skeptical of press releases. Nothing on this show is financial advice.
Also from Brian Lampert: Concrete Compute and Space Stakes.
Britain's financial watchdog has set up what is billed as a global regulator first to test whether quantum computers can catch money launderers and fraudsters better than today's systems, with firms invited to join by 31 December 2026 for a launch in Q1 2027. The twist is the hardware it would need is not ready yet, and banks have real money riding on whether early tests point anywhere useful. How do you run a fair test for an advantage that does not exist yet? That's what today's episode will work out. Welcome back to Quickly Quantum, your daily brief on the quantum frontier. It's Monday, November 9, 2026. Let's get into it.
Let's start with the idea you will need for the rest of today's show, because the day's biggest story turns on it. Quantum advantage is one of those phrases you hear a lot, so what does it actually mean when physicists use it? In plain language it means a programmable quantum computer solving a defined problem that no classical computer can solve in any feasible amount of time, even if that problem itself is not yet useful. That last clause matters for you, because a demonstration can count as an advantage in the scientific sense while changing nothing for a bank or a lab the next morning. Conceptually it always involves two jobs at once, the engineering job of building a capable quantum machine, and the complexity job of finding a problem where that machine has a superpolynomial speedup over the best known or possible classical approach for that task. Researchers tend to treat that kind of result as primarily a scientific milestone, with relatively little immediate bearing on commercial viability, partly because classical computers and classical algorithms keep improving in unpredictable ways, which means a claimed lead can be temporary or unstable and deserves scrutiny. A classic family of proposals involves sampling tasks, where you take measurements from processes like boson sampling or random quantum circuits. Those outputs look flat but have a hidden structure that makes them hard to mimic classically without very strong assumptions, which is why sampling became a favored path for early demonstrations. The hard part, and the part that matters for the main story later in today's show, is the word useful. A useful advantage would mean a clearly defined task that matters to someone, a credible comparison against the best classical method for that same task, and a result others can verify. Until those three line up, you have an interesting experiment, and you should hear any bigger claim as a promise being tested, which is exactly the frame to carry with you today.
The day's biggest technical result comes from error correction, which is the machinery that will decide whether quantum computers ever run long calculations you can trust. The core technique, as reported, pairs up known base mechanisms to discover higher-order hyperedges, meaning linked error events that fire together, with an XOR composition of logical effects so each discovery updates the logical outcome correctly, then decomposes those hyperedges so they still run on fast graph decoders such as PyMatching. Applied to Google's below-threshold superconducting experimental dataset, the augmented priors yielded Lambda = 2.02 plus or minus 0.06 across distances 5 to 7, a 9% improvement over baseline reinforcement-learning-optimized priors, and lowered the distance-7 logical error rate by 14%.
A very different kind of coherence result comes from materials, and if you care about devices that work without a giant fridge, you will want to hear this one. According to Quantum Zeitgeist, researchers led by Professor LU Haipeng at The Hong Kong University of Science and Technology's Department of Chemistry built a perovskite chiral multiple quantum-well system, which means stacked ultra-thin layers of a crystal that sorts electrons by their spin, a quantum property that acts like a tiny compass direction, and stretched spin polarization lifetime to approximately 300 times longer at room temperature. The team credits a spin cascade, a continuous handoff of spin information through intermediate polarized states that softens energy-related Rashba splitting at interfaces, a distortion that normally scrambles spins fast. They built it with a two-step seed-assisted crystallization approach to make well-oriented wells with varying widths and intrinsic chirality, creating an energy gradient and fully chiral environment. That is lab-scale optoelectronics, and I see it as intriguing for spin devices rather than a qubit milestone.
From warm materials to molecules that are warm by quantum standards, there is a new entanglement record with a clever trick behind it. According to Quantum Zeitgeist, a team led by John Doyle entangled laser-cooled calcium monofluoride molecules, CaF, with a fidelity of 97.6 (+0.8/-1.1)%, described as the highest reported entanglement achieved directly with laser-cooled molecules. The advance came from arranging molecules in specific three-dimensional geometries, lifting interaction quality by approximately 40% over previous two-dimensional arrangements, plus a geometric echo protocol that rapidly moves one molecule to a mirror-image spot midway through to cancel drift in trap positions and boost coherence by a factor of three. Thermal motion and trap drift normally wreck these delicate interactions, so extending reliable interaction time matters. The striking detail for you is the temperature, because these CaF molecules were substantially hotter than molecules in other leading experiments that cool near motional ground state, yet still made maximally entangled Bell pairs, which are pairs whose outcomes stay linked no matter how far apart you measure them. The team projects refinements could push to 99.9% or higher, and believes the methods apply broadly to other molecular tweezer platforms.
Meanwhile the superconducting world is going back to basics, to a layer only atoms thick. According to edhat.com, three UC Santa Barbara researchers have joined a new institute called MARQUIS: Manufacturable and Resilient Superconducting Quantum Information Systems, to re-engineer atomic-scale Josephson junctions, which are the tiny circuit elements at the heart of superconducting qubits made today from two aluminum layers around aluminum oxide roughly a nanometer thick. Led by Princeton University, the institute will receive $27.9 million from the National Science Foundation over five years and bring together roughly two dozen laboratories across nine institutions, spanning materials science, quantum devices and semiconductor processing. UCSB's team is Nobel laureate Michel Devoret, computational materials specialist Chris Van de Walle and grower of complex materials Chris Palmstrom. Devoret is quoted saying the Josephson tunnel junction is now, finally, the weakest element in a superconducting quantum processor, after decades clearing other roadblocks like electromagnetic noise and wiring losses. That framing matters for you because it says the next gains may come from manufacturing the junction itself, and MARQUIS is one of eight institutes in a $290 million NSF Quantum Leap Challenge Institutes round.
Proven news breaker @QtumAnalysis posted on X that THE U.S. GOVERNMENT AGREED TO BUY GAS MINED FROM THE MOON. MANY QUANTUM COMPUTERS CAN'T RUN WITHOUT IT, according to that post. Proven news breaker @QtumAnalysis posted on X that it is the first time the U.S. government has agreed to buy a natural resource from space, and that superconducting machines like those IBM and Google build need temperatures colder than deep space, so the fridges that get them there run on helium-3. It adds that trapped-ion and neutral-atom machines like IonQ's, Quantinuum's and Infleqtion's do not use these fridges, so a squeeze hits superconducting players hardest. Nothing has been mined yet, the contracts point to the future, and for you the takeaway is supply chain, not science.
And to close the news run on a lighter note you could actually visit, science as magic school is now real in Busan. According to The Quantum Insider, citing Seoul Economic Daily, the National Science Museum in Busan partnered with KRISS to launch its Quantum Magic School exhibition, with interactive demos of superposition, entanglement and uncertainty, alongside a model of a 50-qubit superconducting quantum computer and an actual quantum processing unit. The exhibition runs through Nov. 22, and the Science Field Trip tour is scheduled for the 14th. If you went, you would get the fun version first and the lab version after.
That's the rest of today's quantum news. Now, the story I want to slow down for. Today's story is Britain's markets regulator deciding it should run quantum experiments itself before quantum breaks or fixes finance. According to fintechfutures.com, the UK Financial Conduct Authority has set up the Quantum Experimentation Group, or QEG, to explore how quantum computing could help tackle financial crime, billed as a global regulator first. The QEG will initially evaluate quantum approaches to anti-money laundering, AML, which means tracing dirty money through tangled transaction networks, and fraud detection, which means spotting fake patterns fast. The QEG will "explore applications, build capability, and generate evidence on common regulatory challenges, with a focus on detecting financial crime for the initial use case," the watchdog continues. The FCA says it will define problem statements and work alongside participants to build understanding of opportunities, limitations and practical considerations. Expressions of interest are open to financial services firms, quantum computing providers and research labs, with a deadline of 31 December 2026 and the initiative scheduled to launch in Q1 2027. Jessica Rusu, chief data, information and intelligence officer at the FCA, is quoted saying, "We want to help firms explore where quantum could make a meaningful difference, and early experimentation is essential to understanding its potential in financial services," adding that bringing together firms, technology providers and researchers can build readiness for common industry challenges starting with financial crime. For context only, the same article notes that in August, the US Department of the Treasury formed the Quantum-Readiness Task Force to support the sector's transition to quantum-safe technology in an orderly and operationally resilient manner. That pairing tells you where regulators live right now, one eye on quantum as a detection tool, the other on quantum as a risk to encryption. Quantum machine learning, the idea at the center here, just means machine-learning models that run partly on quantum hardware to find patterns in data like laundering graphs. In theory those models could spot correlations classical systems miss, in practice no one has shown that edge on live bank data. If you are new, quantum advantage is simply a quantum machine beating the best classical machine on a defined task, and yesterday I said I am moderately persuaded by the direction of travel but not by any broad claim it is already useful. What is new today strengthens the need for exactly that test, because a regulator is offering to define the task and the baseline comparison in public.
So does a regulator running the experiment get us closer to a real answer, and what would count as winning? The strongest case for this approach is process, not physics. Banks will not share laundering data easily, classical baselines keep shifting, and vendors tend to grade their own homework, so an FCA-led group that writes down the detection problem, the data and what beating classical means could do something the market has not, create a credible comparison. That is why the Q1 2027 launch and 31 December 2026 deadline matter to you, they turn a vague promise into a dated test with named participants still to come. But which firms and labs join, and what comparison to classical baselines will count as success? The reporting does not say yet, and I will not treat this as enforcement or as proof of advantage, because it is an experimentation program. The honest limits are stark, quantum machine learning for AML remains unproven, the tests will be early-stage, and any near-term work will run on small or simulated hardware far from production fraud systems. I expect the first results to look like careful learning about data plumbing, costs and where quantum does not help, which is valuable but not a headline. My read is that I am glad the FCA is doing this in the open, because a neutral scorekeeper raises the bar for every vendor claim about catching crime with qubits, and I would become more convinced the moment they publish the tasks, the classical numbers to beat and the error budgets. Time for the Hype Check. I give this a 3 out of 10 for substance today, because the strongest point against a low score is that a world-first regulator-led group with a concrete crime-detection focus and a Q1 2027 start is genuinely new institutional signal, but I land at 3 because there are no results, no participants and no proven quantum edge yet. The concrete thing riding on this is the FCA's problem set for money-laundering networks and fraud patterns, and where today leaves it is blank pages with a deadline. If that group hits published benchmarks with classical baselines beaten by clear margins by late 2027, that tells us useful advantage in finance is moving from talk to evidence. If you follow one thing this year, make it today's episode of the broader showdown between detection dreams and device reality, and send it to the colleague who still thinks quantum catches criminals tomorrow. This has been Quickly Quantum, an AI-voiced podcast, created and built by a real human using today's cutting-edge technology. Nothing you heard on this show is financial advice. I'm Brian Lampert, and I'll catch you all tomorrow — take care!
If today's tour from crime-fighting qubits to Moon gas gave you something to chew on, follow the show wherever you listen so tomorrow's brief finds you first.
Watching on YouTube? Hit subscribe, and tomorrow's briefing will be waiting for you. I also host Space Stakes: the business of the new space race, every day. What actually flew, what the contract is really worth, and who has customers. Find it wherever you get your podcasts.