Latest / Quickly Quantum / Ep 3: Quantum's Next Leap: Real Physics or Just a Really Good Pitch?
Transcript
- Three hundred million dollars — for a computer that doesn't exist yet, built by a company that's barely three months old. That's today's headline out of the quantum world, and it's tangled up with something more personal than you'd expect: the security of every password, every bank transfer, every encrypted message you've ever sent. Welcome back to Quickly Quantum, your daily brief on the quantum frontier. It's Wednesday, July 8, 2026. Let's get into it. Today's docket: a fusion-reactor result that moved IBM's stock, a quantum stock having a brutal month, an acquisition out of France, a giant new laser array in Germany, a supercomputer breaking its own record, a fresh government initiative, and an encryption standard built on 1978 math. But the big story is a bet that quantum's timeline to something useful just got dramatically shorter — or so the pitch deck says. Let's start with hardware doing actual chemistry. IBM shares jumped somewhere between two and a half and three and a half percent this week after the company, working with Oak Ridge National Laboratory and the Cleveland Clinic, used a quantum computer to calculate nine different molecular configurations of a molten salt material — the kind of exotic, corrosive fluid that's a leading candidate for cooling and fueling next-generation fusion reactors. According to the report, this is the first time anyone's run this specific kind of molecular calculation on quantum hardware, and it's part of the Department of Energy's Genesis Mission, a push to accelerate breakthrough energy research using next-generation computing. Here's why a stock market moved on a chemistry paper: fusion has a materials problem, not just a physics problem. You can get the plasma hot enough; keeping the machinery around it from dissolving is the harder part. Simulating how these molten salts behave at the molecular level is exactly the kind of problem quantum computers are supposed to be good at — lots of interacting particles, exponential complexity, the stuff that chokes classical supercomputers. Whether this particular nine-configuration calculation could've been done just as well on a laptop with the right algorithm is a fair question nobody's fully answered, but symbolically, it's a proof point IBM badly wanted: a national lab and a major hospital system trusting IBM's quantum hardware for real applied science, not just benchmarks. It also fits neatly into IBM's broader roadmap — the company has talked about roughly ten billion dollars committed to quantum computing, with a fault-tolerant system named Starling targeted for twenty twenty-nine, and a follow-on machine called Blue Jay after that. A materials-science win, even a modest one, is exactly the kind of headline that roadmap needs to stay credible with investors. Meanwhile, in the stock market, it's been a rough stretch for IonQ. Shares fell seven point one percent in a single session, and that's not an isolated dip — IonQ is now down roughly forty-four percent from its October 2025 high, and down around thirty-one percent in just the last thirty days, according to Tickeron. Two things collided here. First, general jitters about quantum stocks trading at what one analysis called extreme price-to-sales multiples — we're talking somewhere in the range of ninety-seven to a hundred and eleven times trailing sales, numbers that make even enthusiastic investors nervous when the broader market gets twitchy. Insider selling hasn't helped sentiment either. Second, and more specifically: European rival IQM Quantum Computers just started trading on the Nasdaq, which means there's a new publicly-traded pure-play quantum stock soaking up investor dollars that might otherwise have gone to IonQ. It's a reminder that quantum investing right now is as much about sector rotation and story-of-the-day as it is about qubit counts. A new listing doesn't change IonQ's technology roadmap one bit, but it changes who's competing for the same pool of speculative capital. Add in IBM's very public quantum push — the same IBM whose stock just popped on that fusion news — and you've got a market that's genuinely uncertain how to price companies whose revenue today is a rounding error compared to the valuations they carry. None of this tells you whether IonQ's trapped-ion technology is good or bad. It tells you the market is still treating quantum stocks like lottery tickets, and lottery tickets get repriced fast when a new one shows up on the shelf. Next, a smaller deal worth a beat of attention if you care about who's quietly building the picks and shovels of this industry. BTQ Technologies has closed its acquisition of QPerfect, a French quantum-software company, for around eighteen point six million euros, with up to another five point seven million euros in earnout payments if certain milestones are hit. What BTQ is actually buying is QPerfect's MIMIQ emulator and Digital Twin software — tools that let you simulate how a quantum computer would behave, on a classical machine, before you ever touch real hardware. Specifically, QPerfect specializes in emulating neutral-atom systems, the same underlying technology — atoms trapped by lasers — that our main story today is built on. That's not a coincidence; neutral atoms are having a moment across the industry right now, and good simulation tools are how you design error-correction schemes and test algorithms without burning expensive machine time. The deal also gives BTQ a new European research hub in Strasbourg, which matters for a company that's been trying to position itself at the intersection of quantum computing and post-quantum cryptography — the encryption methods designed to survive an attack from a quantum computer. Keep an eye on Strasbourg — European quantum research has been quietly building critical mass, and hubs like this are part of why smaller countries are punching above their weight in this field. This is the unglamorous side of the industry: not a breakthrough qubit count, just infrastructure consolidation. But infrastructure consolidation is usually a sign a sector is maturing past its pure-hype phase, one small acquisition at a time. Two more hardware stories, both about brute-force engineering rather than headline claims. In Germany, researchers at Fraunhofer ILT in Aachen built a laser-optical system that splits four laser beams into two thousand individually controllable optical tweezers — think of it as two thousand separate tiny robotic hands, each one able to grab and hold a single atom in place with sub-micrometer precision. That's the kind of unglamorous optical engineering that neutral-atom quantum computers actually depend on, and this particular system is destined for a machine under construction at the University of Stuttgart. It's a good reminder that scaling quantum computers isn't just about clever algorithms — a huge amount of it is old-fashioned laser and optics engineering, done at a level of precision that would've sounded like science fiction a decade ago. On the classical side, researchers at the Jülich Supercomputing Centre and NVIDIA say they simulated a fifty-qubit quantum system on Jülich's JUPITER exascale supercomputer, beating a prior record of forty-eight qubits. Quantum Zeitgeist reported this one, and we haven't independently confirmed it elsewhere yet, so file it as reported rather than settled. If accurate, it's a useful gut-check on the whole field: exascale classical computers — the fastest conventional supercomputers we have — can still brute-force simulate systems in the fifty-qubit range. That number matters because it sets the bar quantum computers actually need to clear to claim a real, defensible advantage. Every time classical simulation pushes that bar up, it's one more qubit a quantum machine has to add before anyone can credibly say a classical computer literally cannot do this. It's not as flashy as a funding round, but it's exactly the kind of scoreboard-keeping that keeps this industry honest. Two more items, both about the plumbing governments are laying down for the quantum era. First, the National Science Foundation announced something it's calling Project Triad — an initiative to fuse three previously separate branches of the field: quantum sensing, quantum networking, and quantum computing, into one integrated system aimed at real-world applications like healthcare. HPCwire covered the announcement on X, describing it as first-of-its-kind. The idea, in plain terms, is that a quantum sensor that can detect incredibly tiny changes, a quantum network that can move fragile quantum information without destroying it, and a quantum computer that can process it, are more useful together than as three separate research programs that never talk to each other. It's early-stage and light on specifics so far, but it's a sign the U.S. government is thinking about quantum as an integrated technology stack, not just a race to more qubits. Second — and this one connects directly to today's main story — the International Organization for Standardization has adopted Classic McEliece as an official international standard for post-quantum cryptography. That's notable because Classic McEliece isn't some brand-new invention; it's based on code-based cryptography research from 1978, math that's been sitting around for decades precisely because nobody's found a way to break it, quantum computer or not. Quantum Zeitgeist reported this one, and again, we haven't independently verified it beyond that single source. But if it holds up, it's a meaningful milestone in the slow, unglamorous work of migrating the world's encryption before a fault-tolerant quantum computer shows up capable of running Shor's algorithm — which, as you're about to hear, is exactly the machine one very well-funded startup says it's building. Our main story today: Oratomic, a Caltech spinout that came out of stealth mode barely three months ago, just closed a three hundred million dollar Series A. Let's sit with that number for a second — three hundred million dollars, in a single early-stage round, for a company that as far as we know has zero commercial product and zero revenue. The round was co-led by ARCH Venture Partners, Spark Capital, and Khosla Ventures — heavy hitters in deep tech — with participation from Bezos Expeditions, Index Ventures, General Catalyst, and a list of other funds. And notably, theoretical computer scientist Scott Aaronson, one of the most prominent and famously skeptical voices in quantum computing, is listed among the backers personally. When someone who's spent years publicly poking holes in overhyped quantum claims puts his own money in, that's worth noting. So what is Oratomic actually claiming? The company is building neutral-atom quantum computers — machines that trap individual atoms using tightly focused lasers, sometimes called optical tweezers, and use those atoms as qubits, the basic unit of quantum information. The pitch, in plain English: building a fault-tolerant quantum computer — one that can actively correct its own errors as it runs, rather than being knocked over by the tiniest disturbance — has long been assumed to require millions of physical qubits, because you need a lot of imperfect qubits ganged together to create one reliable, error-corrected 'logical' qubit. Oratomic's research, developed with Caltech scientists, argues that a new, more efficient error-correction scheme could cut that requirement down to somewhere between ten and twenty thousand qubits. If true, that's a reduction of multiple orders of magnitude — the difference between a machine that might exist this decade and one that might not exist in our lifetimes. Co-founder Manuel Endres, a Caltech physicist, has already demonstrated laser-trapped arrays of around six thousand atoms — not yet a working fault-tolerant computer, but a real experimental data point suggesting the scale isn't pure fantasy. CEO Dolev Bluvstein put it this way: quote, "Our new research advances simultaneously changed all of our minds." End quote. He's said the founding team — which includes researchers from Caltech, Berkeley, Harvard, Amazon, and Google, along with well-known names like John Preskill — had previously believed commercially useful quantum computing was far away, and this specific error-correction result is what flipped them. The money is earmarked for expanding hardware fabrication, growing the physics and engineering teams, and pushing forward algorithmic research, including cryogenic chip packaging work aimed at chemistry and machine-learning applications down the road. And this raise doesn't exist in a vacuum — it lands during what's reportedly been a record year for private quantum investment. Which raises the obvious question: is this genuine technical conviction, or is this the quantum industry's version of a very hot funding market finding its next big story to chase? Let's get into what would have to be true for Oratomic's bet to actually pay off. So here's my read. The team is genuinely elite — Endres, Preskill, researchers pulled from Caltech, Berkeley, Harvard, Amazon, and Google — and the science underneath this isn't nothing. Neutral-atom systems really are one of the more promising paths in the field right now, precisely because you can rearrange atoms mid-computation, which gives you more flexibility for wiring up error correction than fixed architectures. But — — let's be honest about what's actually been proven versus claimed. The ten-thousand-qubit number rests on a single arXiv paper, and one outlet covering this story flagged that turning that math into working silicon-and-laser hardware is, quote, "a step Oratomic has yet to demonstrate." The company's own announcement is self-reported, straight off its own blog. And the target for an actual fault-tolerant machine is the end of the decade — which, conveniently, is a timeline with no near-term milestone anyone can hold them to. Bluvstein himself has said it's, quote, "plausible, although not guaranteed," that they hit that window. That's an honest hedge, credit where due, but it's also a long way from proof. On X, the reactions split predictably along who's writing the check. Spark Capital co-lead investor @santopoliti posted on X that Oratomic reminds his firm of its partnership with Anthropic — quote, "a rare team, at a rare moment, turning a decades-old dream into a practical engineering plan." Spark Capital's own account posted on X calling Oratomic's aim, quote, "the world's first fault-tolerant quantum computer using neutral atom qubits" — though notice that's a claim about what they're building, not what exists yet. And early backer @FinnMurphy12, a GP at Nebular, wrote on X that if any team is going to make quantum computers useful and ubiquitous, quote, "it is Oratomic." That's investors talking their book, which is exactly what you'd expect — nobody puts three hundred million dollars behind a company and then goes on X to hedge. The more useful voice here might be @paullecoque, an investor who drew the contrast with public quantum companies like IonQ. He wrote on X that Oratomic's Caltech research on lowering the qubit threshold for fault tolerance is, quote, "worth taking note of," but stressed it's, quote, "crucial to separate promising science from proven execution," pointing out the company is just over three months old, still lab-stage, pre-revenue, with no commercial product. That's the sentence I'd tattoo on this whole story. Then there's the part that should genuinely concern you even if you don't own a single crypto wallet: Oratomic is explicitly framing its roadmap around Shor's algorithm — the quantum algorithm that can break the public-key encryption protecting basically everything online. That's not incidental marketing; it's their stated purpose, and it's why this raise is landing in the same week as broader coverage of encryption teams bracing for so-called Q-Day. Governments are already migrating to post-quantum cryptography, with many plans targeting completion by 2035. If Oratomic's timeline is even directionally right, that migration deadline suddenly looks less like caution and more like a photo finish. Time for the Hype Check. I'm putting this one at a five. The team, the physics pedigree, and the funding are all real and serious — this isn't vaporware from nobodies. But the headline number is unproven math from a company with no hardware yet built to demonstrate it, on a self-reported timeline with zero falsifiable checkpoints before the end of the decade. Real science, real money, real uncertainty — a five feels right, and I'll happily revise it up the day they show working error-corrected qubits instead of a paper. If you're the kind of person who reads a three-hundred-million-dollar funding announcement with one eyebrow raised — good, that's basically the job description for this show, and I'd love to have you back tomorrow. Follow Quickly Quantum wherever you're listening right now, so the next episode just shows up. That's Quickly Quantum for today. New episodes every day. This is an AI-voiced podcast, created and built by a real person using today's cutting-edge technology. And remember: nothing on this show is financial advice. I'm Brian Lampert — see you tomorrow.