Latest / Quickly Quantum / 120-Qubit Links, and Intel’s Quantum Stack Bet — October 27, 2026
Transcript
- Intel has entered into an agreement to acquire Classiq, a quantum software company, in a move intended to make Intel a full-stack quantum computing provider, according to Quantum Computing Report. The complication is that owning software and hardware on paper doesn’t establish how well they’ll work together at scale, and the deal has not closed. Intel says it will keep Classiq’s platform hardware-agnostic after the deal closes, a meaningful promise for developers who want software to work across different machines. What should you make of a software-and-hardware deal when the machines are still being developed? Today we’ll sort the strategic bet from what the agreement actually demonstrates. Welcome back to Quickly Quantum, your daily brief on the quantum frontier. It's Tuesday, October 27, 2026. Let's get into it. IBM researchers connected two 120-qubit processors using four superconducting cables, and the links transferred quantum states with an average fidelity of 99%, The Quantum Insider reports. Fidelity is a measure of how closely a result matches the intended one. The researchers also performed operations and created entangled states across the two chips; entanglement is a shared quantum state between separate qubits. That’s the practical point: a future machine could be built from connected modules rather than relying on one enormous chip. The work appeared as a preprint on arXiv, and the result has an important boundary: complete operations between processors averaged 94.75% fidelity, below the state-transfer result. Only selected qubits near the connections were wired and measured, so the experiment did not coordinate all 240 qubits. The team used four aluminum coaxial cables, each about 80 centimeters long, inside the system’s ultracold environment. The researchers describe this as, to their knowledge, the first multichannel quantum interconnect between large superconducting processors. We’ve been careful about fidelity headlines: a gate result and a state-transfer result measure different things. This one is a new link between processors, and the lower figure for complete operations is part of the same story. I think the more useful signal is that the team got multiple connections working between large processors; the next challenge is getting more of the machine involved. Meanwhile, IBM’s experiment is about linking machines, while a UK program is testing what it takes to put them somewhere people can use them. The Quantum Insider reports that Britain’s £30 million quantum computing testbed program helped seven suppliers deploy prototypes outside their own laboratories, across five approaches to quantum computing. The National Quantum Computing Centre’s Phase 1 report says suppliers had to tackle installation, operation and customer support alongside hardware development. Six testbeds entered independent technical evaluation, and the main testing program is due to conclude in the second quarter of 2027. The report says collaborative projects secured £4.6 million in follow-on funding. This is a program evaluation, not an independent ranking of which machine performs best. Among the examples, Quantum Motion said preparing its QM One silicon-spin system for external users pushed its work from individual components toward engineering a complete computer, including software for running circuits. I like the emphasis here: a machine has to be delivered and operated before a customer can learn what it’s good for. From government-backed testbeds to government-backed security, IDEMIA and South Korea’s KT have moved quantum-resistant SIMs from testing to deployment in military networks, Quantum Computing Report says. The work sits within South Korea’s 2026 Post-Quantum Cryptography Pilot Conversion Project. The SIMs support international NIST standards alongside Korea’s domestic KpqC algorithms; post-quantum cryptography means encryption designed to resist attacks from future quantum computers. The article describes a secure element inside the SIM as a hardware root of trust, a protected place to run cryptographic operations. The pilot also tests whether the system can switch between standards without rewriting application software, a capability called crypto-agility. This is a pilot deployment, not a force-wide rollout. For you, the near-term significance is migration work happening in actual defense communications, while the scale and certification details remain unclear. The useful step is deployment; the claim is still bounded by the pilot. And in another security story, the proposed infrastructure is in orbit. Quantum Zeitgeist reports that SEALSQ signed a binding letter of intent to acquire satellite capacity and operational services from twelve WISeSat Space satellites, in a transaction estimated at $15 million over the next 18 months. The plan is to establish initial infrastructure for SEALSQ’s Quantum Spatial Orbital Cloud, intended for post-quantum-secured satellite communications and digital identity. The initial fleet, together with a ground station, is intended to support integration of semiconductor security, digital identity and cryptographic-key management across the network. SEALSQ’s CEO Carlos Moreira called the letter of intent ‘a concrete step towards moving QSOC from vision into infrastructure.’ That’s his characterization, and the article describes the next work as integration and validation before customer services can begin. I’d treat the distinction as central: this is an agreement to acquire capacity, with the proposed services still to be developed and validated. A different kind of quantum infrastructure story comes from a public utility in Tennessee. Proven news breaker @techinnovationz posted on X that EPB, Chattanooga’s municipal utility, operates a fiber network and now sells quantum computing time to the public, and that IonQ describes EPB as the first utility in the world to operate both a commercial quantum network and a quantum computer. The post says the network has been commercial since 2023 and the IonQ Forte Enterprise has been live since September 18; those details come from the post, not independent confirmation here. It also says the University of Tennessee at Chattanooga is the first customer, eight fellows funded by a $4M NIST grant are running grid algorithms, and IonQ’s $15M, five-year research center will host a commercial quantum memory on live fiber. The post presents the fiber network as a foundation for IonQ’s plan to link quantum computers. There’s a concrete local story in that account: a utility building an existing network and pairing it with quantum services. The results will matter more than the ambition. That's the rest of today's quantum news. Now, the story I want to slow down for. Intel is trying to pair its quantum hardware work with software that helps people design quantum programs. Quantum Computing Report says Intel has entered into a definitive agreement to acquire Israeli quantum software provider Classiq, with the combination designed to make Intel a full-stack quantum computing provider. A full stack means the company is bringing together the machine and software used to program it. Classiq’s platform automates algorithm synthesis, turning a high-level description of a quantum task into circuits that can run on hardware. That matters because developers otherwise have to work through detailed circuit design and hardware constraints themselves. Intel has been developing quantum capabilities since 2015, according to Quantum Computing Report, including silicon spin-qubit chips and cryogenic control electronics. Silicon spin qubits store information using the spin of electrons in silicon; cryogenic control electronics operate in extremely cold conditions needed by the system. Classiq brings a software platform intended to abstract some of the complexity of circuit design. The comparison in the article is to electronic design automation and computer-aided design, software that helps engineers design conventional chips and other products. Intel CEO Lip-Bu Tan said, “Joining forces with Classiq transforms Intel’s Quantum Computing Group into a full-stack quantum computing business,” and pointed to the complementary nature of Intel’s manufacturing scale and Classiq’s software innovation. Intel’s own post on X says it has entered into an agreement to acquire @ClassiqTech to advance scalable quantum computing. Those statements describe the company’s intended direction; the agreement itself is still subject to customary regulatory approvals. The software side is the strategic hinge. A quantum processor can exist without being easy for developers to use, and software that simplifies programming could make a hardware platform more accessible. Classiq’s reported approach is hardware- and modality-agnostic, meaning it is designed to work across different kinds of quantum hardware. Quantum Computing Report says Intel has committed to maintaining that approach, and that after the deal closes it will continue supporting Classiq’s existing multi-platform integrations. That promise gives the deal a broader reach than simply attaching one software layer to Intel’s own machines. But the hard part is making the pieces work together in practice. Quantum Computing Report says Intel and Classiq will continue to operate as independent entities until the transaction closes, and the deal remains subject to regulatory approvals. The article says the transaction is expected to close in the first half of 2027. That timing is an expectation, not a completed acquisition. The reporting also leaves us with a distinction between the business move and a technical result. This agreement doesn’t itself demonstrate a new quantum processor, a better error rate or a useful application. It does put Intel’s manufacturing ambitions beside Classiq’s software, and it sets out an intention to preserve Classiq’s integrations across hardware platforms. Those are meaningful pieces of strategy, but they are not evidence yet that the combined business can scale a useful quantum computer. Bloomberg’s account, posted by @business on X, described Classiq as an Israeli startup and called quantum computing “a still-nascent field that is seeking commercial applications.” That’s a fair reminder of the commercial challenge. The acquisition can strengthen Intel’s position in the software layer, but the question of who will build valuable applications remains open. I think this is a consequential consolidation move because the software that makes hardware usable can shape which platforms developers choose. That’s my read of the strategic logic, not proof that Intel has secured the winning stack. I’m also encouraged that Intel says it will preserve Classiq’s cross-platform approach; if that commitment survives integration, it could keep the software useful beyond Intel’s own hardware. The open point is whether the combination can deliver practical value as the machines mature. Evidence of working applications and a clear fit between the software and Intel’s hardware would strengthen my view; stalled integration would weaken it. Time for the Hype Check. I’d give this a six out of ten for substance: the strongest point against that score is that this is an agreement, not a demonstrated technical advance, but the named software-and-hardware combination and the stated plan to maintain multi-platform support make it more than a vague quantum promise. The stake is Intel’s place in the quantum stack: if this integration works, Intel and Classiq gain a broader platform position; if it falters, both lose time and strategic ground while the field is still searching for commercial applications. If this kind of clear-eyed quantum coverage is useful, follow today’s episode wherever you listen. 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! Watching on YouTube? Hit subscribe, and tomorrow's briefing will be waiting for you. 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.