Latest / Quickly Quantum / August 26, 2026 — Dead and Alive: Quantum Computing's Credibility Gap
Transcript
- Valuations running as high as three thousand times sales — for quantum computing companies that aren't yet profitable. That's the number sitting under today's main story: is quantum computing actually arriving, or is Wall Street pricing in a future the hardware hasn't earned yet? Today on Quickly Quantum: Bloomberg's big new piece asks whether the dream of quantum computing is dead or alive — and the honest answer might genuinely be both. Before that, in the headlines: NIST widens a key photon detector by a hundred times, Sweden joins the national quantum strategy race, D-Wave's longtime CFO heads for the exit, IonQ re-runs its energy-efficiency numbers, and a live fight breaks out over which encryption method institutions should trust. Welcome back to Quickly Quantum, your daily brief on the quantum frontier. It's Wednesday, August 26, 2026. Let's get into it. Let's start with a genuine engineering win, not a press-release claim. Superconducting nanowire single-photon detectors — SNSPDs, the ultra-sensitive light counters that quantum networks and photonic computers rely on to catch individual photons — have had a stubborn problem: make them wider, and they stop working reliably. NIST physicists just published a fix in the peer-reviewed journal Optica: a magnetic shielding architecture that lets these detectors scale up to a hundred times wider than standard designs without losing performance. That matters because detector width has been a real manufacturing chokepoint for the photon-counting systems used in quantum networking and photonic quantum computing. This is peer-reviewed hardware work, which is rarer than it should be in this field. The catch: lab-scale fabrication improvements don't automatically show up in deployed network hardware, and NIST hasn't given a commercialization timeline. Still, if you're building a photonic quantum network, this is exactly the unglamorous plumbing fix that determines whether your grand architecture actually ships. Sweden just became the newest country to formalize a national quantum strategy, this one running through 2036 and covering everything from academic-to-commercial funding pipelines to defense resilience. The plan leans on the Wallenberg Centre for Quantum Technology at Chalmers and the newly formed Swedish Centre for Quantum Technology, and it directs public institutions and critical infrastructure operators to start migrating to post-quantum encryption ahead of European Union deadlines running through 2030. It also folds into Nordic and transatlantic partnerships, including a U.S.-Sweden technology agreement and NATO's own quantum efforts. If you're tracking the sovereign-quantum race, add Sweden to a growing list that already includes the U.S., the U.K., Germany, and India — all chasing the same thing, which tells you this stopped being a pure technology story and became a geopolitics story a while ago. My skeptic note here is the one that applies to basically every national strategy document: publishing a strategy isn't the same as funding it. The real test is whether specific budget lines show up behind these pillars. Sticking with leadership moves — D-Wave's chief financial officer, John Markovich, is retiring effective September second, after five years with the company. In that stretch he helped take D-Wave public back in 2022 and helped raise over nine hundred million dollars in capital. The company is explicit this isn't tied to any accounting dispute or business disagreement — Greg Golkov, D-Wave's SVP of finance, who's spent more than two decades in finance roles including at Butterfly Network, steps in as acting CFO. On its own, a CFO retirement doesn't move markets. But D-Wave is one of the few publicly traded, dual-platform quantum companies — it sells both annealing systems and gate-model hardware — and any executive transition at a small-cap stock with a sky-high valuation draws extra scrutiny almost by reflex. If you own D-Wave stock, this is worth a beat of attention, not panic. So how much of this is genuinely new? IonQ posted fresh numbers this week claiming its thirty-six-qubit Forte Enterprise system beats classical GPU simulation on energy use for AI fine-tuning — real measured hardware, they say, not simulation, with accuracy gains up to twenty-four percent past roughly thirty-four qubits. That's according to IonQ's own account on X, posted by @IonQ_Inc, crediting researcher Oliver Knitter and team. Here's the thing though — this isn't a new experiment. The Quantum Insider reported essentially the same result, same Forte Enterprise system, same roughly-thirty-four-qubit energy break-even point, back in July, using direct power monitoring instead of theoretical estimates. So this looks like IonQ re-amplifying a summer result with a fresh accuracy figure attached. And IonQ's own researchers, along with outside analysts at Futurum Group, caution the crossover point depends heavily on the specific hardware, GPU, and circuit design used — it's one workload, not a general quantum advantage. Now here's a live argument worth eavesdropping on — and why should a crypto founder's podcast appearance matter to a quantum show? Because the standards fight happening in Web3 right now is the same fight playing out everywhere post-quantum migration gets discussed. NEAR's official account, @NEARProtocol, posted the clip on X, and according to X's own view count on that post, it's pulled north of a hundred thousand impressions — though that's the platform's own metric, not something we've independently verified. Replies push back — hash-based PQC proponents favor those schemes because their security assumptions are older, simpler, and more thoroughly stress-tested, even if they're clunkier for some use cases. Worth remembering this is one founder's stated preference, not industry consensus — NIST itself has finalized multiple algorithms from both families rather than crowning one. But that fight happening this week, the same week federal post-quantum migration timelines are in the news, tells you the harvest-now-decrypt-later clock feels real to the people who'd actually get hurt by it. And that tension — real cryptographic progress meeting still-unresolved questions about which path wins — is exactly what today's main story digs into, just one level up. Our main story today — call it quantum's credibility gap, because that's really what's being litigated in a big new piece from Bloomberg's Mark Bergen, published this morning. The piece asks the question that's been quietly hanging over this whole field for years: is quantum computing actually here, or is the industry's storytelling running way out ahead of what the machines can do? Bergen's own framing is that the field is genuinely both — dead and alive at the same time. Before you write that off as a cute headline trick, let's walk through why that duality is actually earned, because both halves of it are backed by real, current evidence. Start with the alive half. Quantinuum and IonQ are now reporting two-qubit gate fidelities — how reliably a quantum computer performs its basic two-qubit operation without an error — above ninety-nine point nine percent: Quantinuum at ninety-nine point nine one four percent, IonQ at ninety-nine point nine nine percent, both clearing the threshold most researchers agree you need before error correction actually starts paying off instead of adding more noise than it removes. Google's Willow chip is close but hasn't cleared that same bar — its publicly reported two-qubit fidelity sits around ninety-nine point eight five to ninety-nine point eight eight percent. So the field genuinely has multiple platforms pushing toward fault-tolerance-grade numbers, even if they're not all there yet. And it's moved some of the field's most credentialed skeptics. Scott Aaronson — the theoretical computer scientist who's spent two decades as quantum computing's most influential public skeptic — has now said it's a live possibility that a fault-tolerant machine, meaning one actually correcting its own errors, running Shor's algorithm, the code-breaking algorithm that threatens modern encryption, could exist before the next U.S. presidential election. And DARPA's own program lead, Micah Stoutimore, has said publicly it now seems likely someone builds a utility-scale quantum computer — one actually useful for real problems, not just demonstrations — by 2033. That's the alive case, and it's coming from people who don't hand out optimism casually. Now the dead half. Quantum stocks have been trading on narrative more than revenue, with some estimates putting valuations anywhere from a hundred fifty times to three thousand times sales, for companies that in most cases aren't profitable and won't be for years. Short-seller Wolfpack Research has accused IonQ specifically of masking how much of its revenue depended on Pentagon earmarks that later got canceled. That's an allegation, not a settled fact, but it's exactly the kind of scrutiny that shows up when a stock price and a company's actual delivered product start to diverge. That gap between promise and delivered hardware is the throughline running under basically every quantum stock story this year. So which half do you believe? Honestly, that's the wrong question — and figuring out why is the point of the next few minutes. So let's stack these voices up, because right now this reads like two completely different industries wearing the same headline. On the bull side, you've got Aaronson — a researcher who built his entire reputation on shooting down overhyped quantum claims — now saying a pre-election Shor's-algorithm demonstration is plausible. That's not nothing. When your harshest in-house critic starts moving his own goalposts closer, that's real information. Add DARPA's Micah Stoutimore putting an actual year, 2033, on utility-scale quantum computing, and you've got two independent, credentialed voices converging on timelines that would have sounded like science fiction three years ago. On the skeptic side, you've got Wolfpack Research's allegation that IonQ dressed up its revenue with Pentagon contracts that later got canceled — and you've got valuations running from a hundred fifty times sales up toward three thousand times sales, in a sector where almost nobody's turning a profit yet. Three thousand times sales isn't a valuation, it's a hope wearing a ticker symbol. So how do you square hardware that's actually improving with valuations that don't add up? Here's the honest complication, and it's one Bloomberg's own framing gets right: these two stories aren't actually in conflict. The hardware can be getting genuinely better — fidelity numbers are measured and published, they don't lie — while the business case for that hardware remains almost entirely unproven at commercial scale. Improving qubits and improving quarterly revenue are different problems on different timelines, and this industry has spent years letting investors conflate them. That's not a contradiction in the reporting. That's just the actual state of the field right now. I'll be honest — I've watched this exact duality resurface in quantum coverage for years without it ever fully resolving, real fidelity gains on one side, unproven commercial demand on the other — and today's reporting doesn't break that pattern, it reconfirms it. What would change my mind? A quantum company posting revenue growth that tracks its valuation instead of its press-release cadence. Until that happens, I read the fidelity numbers as genuinely exciting science, and I read the stock prices as a mostly separate conversation — one that has more to do with narrative momentum than with anything running inside a dilution refrigerator today. Here's the split that actually matters: if you're a researcher or an engineer, this is a genuinely good year — the error rates are real, and the Aaronson and DARPA statements are real signals from people with no reason to inflate them. If you're an investor pricing a stock at three thousand times sales, you're not betting on this year's hardware. You're betting on 2033 arriving on schedule, and on your specific company being the one that gets there first — which is a much riskier bet than the chart implies. Time for the Hype Check. I'm putting this one at a 6. The hardware progress — the fidelity numbers, the Aaronson and DARPA statements — is genuinely substantive and independently sourced. But the sector this story describes still leans hard on narrative to justify valuations the revenue doesn't back up yet. Real science, six out of ten honest about what it means for your bank account. If today's split-screen story hit differently depending on whether you code or you invest, that's kind of the point — this show exists to hold both readings at once. Follow Quickly Quantum wherever you listen, and if you know someone who keeps asking whether quantum computing is real, send them today's episode. 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!