Latest / Quickly Quantum / Quantum assist inside crash simulation, Swiss post-quantum chip plan — September 21, 2026
Transcript
- Can you slip a quantum calculation inside the simulation engineers already run and get answers back sooner? That's the test in today's lead story on LS-DYNA, and before that in the headlines, a Swiss post-quantum chip proposal with real money attached, plus a bank-tech board addition at D-Wave and a Chicago reality check. Welcome back to Quickly Quantum, your daily brief on the quantum frontier. It's Monday, September 21, 2026. Let's get into it. Full plate today, so let's move fast. Switzerland wants a sovereign home for chips that can survive quantum attacks. Reported by The Quantum Insider, SEALSQ, WISeKey and the Canton of Jura, a regional government in Switzerland, signed a Memorandum of Understanding, which is a formal agreement to explore a project, to collaborate on the establishment of a Post-Quantum Semiconductor and Cybersecurity Center in the Canton of Jura. The proposed Jura Center represents an indicative investment of approximately CHF 40-60 million over a period of six years, and those words indicative and proposed matter, this is a plan, not a funded fab pouring concrete. Now, what would the site actually do? The Jura Center is expected to establish capabilities for cryptographic root-of-trust injection, secure firmware personalization, testing and certification, complemented progressively by ASIC design capabilities, meaning custom chips built for secure jobs like identity and defense electronics. At the heart of the project will be SEALSQ's QS7001 Quantum Shield, a secure semiconductor designed to implement post-quantum cryptographic algorithms standardized by the U.S. National Institute of Standards and Technology, including ML-KEM and ML-DSA, the new math meant to stay safe when large quantum computers arrive. So who wins if this happens? Your angle is jobs and leverage, with the reporting pointing to approximately 40 direct jobs within two years, 150 by Year 5 and more than 250 by Year 8, with at least 60% expected for Canton residents. @dwavequantum posted on X: "Today, D-Wave announced the appointment of veteran financial technology executive Bernard Gavgani to its Board of Directors and Cybersecurity Committee." Why should you care about a board seat? Because it signals where D-Wave thinks the customer conversation lives right now, enterprise IT, security reviews, and AI governance, not just qubits in a lab. Now, do we know what he will change? We don't, and the post doesn't promise product or revenue effects, so my read stops at strategy signal, this looks like credibility-building for regulated buyers. Chicago says you are mixing up two different machines. In a Chicago Tribune editorial titled Editorial: Quantum is not AI, folks, and Chicago is not building a lakefront data center, the piece contends local quantum plans are being misread as an AI data-center build. For a first-time listener, the distinction is simple, AI systems learn from data and guzzle power at scale, while quantum computers probe physics in a different way to solve select hard problems, and a quantum lab does not look or hire like a warehouse of GPUs. So why does the confusion hurt? Because land use, power draw, and job promises get scored with the wrong math when you swap one for the other. Still, it tees up our lead perfectly, what does useful quantum look like when it has to earn its keep inside real work? Our lead today comes from scientific-computing.com, and candor first, we could not fetch the full article, so we are working from the reported gist. The thesis I want to test in today's episode is this: the quiet quantum wins will look like faster engineering, not bigger computers. Here's the claim and nothing beyond it, IonQ and Ansys work embedded a quantum step in LS-DYNA simulation to cut engineering runtimes. Let me make that land for everyone. IonQ builds quantum computers from trapped ions, charged atoms held in place and steered with lasers, that's your one-sentence newcomer guide. Ansys sells the simulation software engineers use to crash-test cars and planes on a computer before bending metal. Now, LS-DYNA is one of those heavy-duty solvers, the engine that chews through forces, materials, and crashes, and runtimes there mean hours on big clusters where time is literally money. So the idea is you keep the familiar workflow and hide a quantum assist inside, which means low friction for you if it repeats. Now, what don't we know? A lot, and I won't fill gaps on air. Today's fetch gives us no test models, no wall-clock savings, no qubit counts, and no word on whether results hold across customer models, so I will not imply a universal speed-up beyond tested models. Could this be a narrow demo that fades on messy real meshes? Absolutely, that question decides the whole story. That modesty is actually the point, and here is where I land after sitting with the thin sourcing. My view, clearly my opinion, is that hybrid inside LS-DYNA is exactly the right shape for near-term value, small quantum assist, big classical host, no rip-and-replace, which means adoption could move fast if the speed-up repeats and survives queue time, cloud cost, and validation. But shape is not proof, and on one sentence from one feed I cannot call this proven, nor can I tell you which solver stage got the assist or what baseline it beat. Now, what would raise my confidence? The full paper with model sizes, baseline hardware, timed runs, repeat statistics, plus a customer running their own model and saying it held. And why am I still paying attention despite the thin file? Because engineering simulation is where minutes compound into money, and a repeatable shave there would matter sooner than any qubit record. Time for the Hype Check. I give this a 4 out of 10 on substance for today, not because the idea is weak, I like the idea, but because all we can verify on air is that single reported line. A higher score needs numbers we can check. The challenged belief here is that quantum must replace supercomputers to matter, what replaces it is quieter, quantum earning rent as a plug-in that helps the supercomputer finish sooner. If today's episode helped you sort benchmarks from breakthroughs, follow Quickly Quantum wherever you listen so you get tomorrow's brief without hunting. 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! I also host Concrete Compute: a daily briefing on the AI buildout. The datacenters, the megawatts, and who actually pays for them. Find it wherever you get your podcasts.