<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:media="http://search.yahoo.com/mrss/"><channel><title>Explore Quantum</title><description>The editors of Explore Quantum: an AI news platform that writes in English about quantum technology, from facts in international trade media, publications by companies and research institutes and public sources.</description><link>https://explorequantum.com/</link><language>en</language><item><title>Cloud Benchmark Finds Quantum Chips Keep Only a Fraction of Their Qubits Entangled</title><link>https://explorequantum.com/article/cloud-benchmark-exposes-qubit-entanglement-gap/</link><guid isPermaLink="true">https://explorequantum.com/article/cloud-benchmark-exposes-qubit-entanglement-gap/</guid><description>A Spanish-led consortium found the best superconducting chips among nine cloud quantum processors held only 27 and 24 qubits entangled, a small share of their registers. A separate preprint proves new families of maximally entangled states tied to prime-power dimensions.</description><pubDate>Mon, 16 Nov 2026 08:29:32 GMT</pubDate><content:encoded>&lt;p&gt;According to the study, nine cloud-accessible quantum processors from five vendors keep only a sliver of their qubits entangled at once. A Spanish-led consortium of Gradiant, the University of the Basque Country, the Basque Center for Applied Mathematics and several partner institutions ran the tests. The team worked entirely through remote cloud-access services, with no error mitigation. They searched for the largest Greenberger-Horne-Zeilinger (GHZ) state — a maximally entangled state spanning many qubits at once — that each processor could sustain above a genuine-multipartite-entanglement fidelity threshold.&lt;/p&gt;
&lt;p&gt;The gap was stark. According to the study, the best superconducting processors reached GHZ states of only 27 and 24 qubits. Each figure is less than a quarter of the chip’s full register. A 12-qubit trapped-ion processor, by contrast, held its entire register entangled. The study also found that median calibration figures poorly predict which processor will pass. Requiring certified entanglement, rather than raw state size, cuts the number of passing devices from nine down to four. The consortium posted its full method and results as a preprint, with Sergio Hernández and Mikel Sanz among its 13 authors, according to the study. It also published calibration data and measurements in a Zenodo archive.&lt;/p&gt;
&lt;h2 id=&quot;a-separate-theory-result&quot;&gt;A separate theory result&lt;/h2&gt;
&lt;p&gt;A second, unrelated paper posted this month tackles a different open problem in multipartite entanglement, according to the preprint. Palash Pandya describes a method for constructing absolutely maximally entangled (AME) states, a maximally scrambled form of entanglement in which every subset of the particles is as entangled as possible. The method builds such states in local dimension 2q, where q is an odd prime power, by coupling pairs of qubits to q-level quantum systems.&lt;/p&gt;
&lt;p&gt;Dimension 2q admits 13-party AME states for every odd prime power q of 17 or more, according to the preprint. A general bound extends this to groupings of three or more parties once q passes a threshold of max(N, 2 to the power N-2). The same kind of result holds for 9 and 11 parties once q reaches 5, and for 15 parties once q reaches 9. For nine parties specifically, the paper goes further, according to the preprint. It proves AME states exist in every dimension of 3 or more, not only prime-power ones, completing that case in full.&lt;/p&gt;
&lt;p&gt;That completeness comes at a cost. According to the preprint, the bound grows exponentially. A 15-party state needs a local dimension of at least 9, while a 17-party state needs at least 67. The construction gets rapidly more expensive as more particles are added. The result was posted to arXiv on 12 November 2026 and revised the next day, with Pandya as sole author, according to the preprint.&lt;/p&gt;
&lt;p&gt;The benchmark consortium’s Zenodo archive and Pandya’s preprint are both now public, open for other groups to test their own processors and proofs against.&lt;/p&gt;
</content:encoded><dc:creator>Mara Quinlan</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-Konputagailu_kuantikoa_Ikerbasque_ZuSL72.jpg" medium="image"/><category>computing</category></item><item><title>A Neon Film Cuts Superconducting-Qubit Energy Loss, While a Separate Method Senses Hidden Qubit Noise</title><link>https://explorequantum.com/article/neon-film-cuts-qubit-energy-loss-weizmann-noise-sensing/</link><guid isPermaLink="true">https://explorequantum.com/article/neon-film-cuts-qubit-energy-loss-weizmann-noise-sensing/</guid><description>A preprint reports that a solid neon film cuts energy loss in superconducting qubits, while a separate Weizmann Institute preprint describes a method for detecting hidden qubit noise without disturbing the qubit.</description><pubDate>Sun, 15 Nov 2026 14:08:49 GMT</pubDate><content:encoded>&lt;p&gt;A thin film of solid neon cuts one of the main causes of qubit energy loss, according to the preprint authors, physicist Palmer and colleagues. The film is deposited onto superconducting circuits and then evaporated away. The loss comes from two-level systems, microscopic defects that trap microwave energy and shorten how long a qubit holds its state. Palmer’s team measured the average internal quality factor of superconducting resonators before and after the treatment. It improved by about 4.3 times on one device and about 2.4 times on another. At higher microwave powers, the gain grew to more than an order of magnitude.&lt;/p&gt;
&lt;p&gt;The preprint authors tested two niobium titanium nitride coplanar-waveguide resonators. One carried a 40-nanometre neon film on high-resistivity silicon, the other a 12-nanometre film on sapphire. Depositing and then evaporating the neon redistributes surface contaminants that otherwise feed two-level-system losses, the authors say. That makes it a non-destructive alternative to cleaning methods that can damage sensitive superconducting materials.&lt;/p&gt;
&lt;p&gt;The quality-factor gains, the preprint authors report, did not disappear once the neon film was gone. They held up after the sample went through a full thermal cycle. The team saw similar behaviour across resonator samples measured in different dilution refrigerators. The size of the improvement still varied with each sample’s surface characteristics.&lt;/p&gt;
&lt;p&gt;A second, unrelated preprint tackles a different obstacle for superconducting qubits, the researchers say: catching the noise a qubit generates without disturbing the qubit to measure it. Nitzan Kahn and colleagues at the Weizmann Institute of Science report an upper bound on that noise. The bound is about 0.70×10^3 rad²/s on the intrinsic frequency-noise power spectral density of a superconducting qubit, measured at a qubit-cavity detuning of 508 megahertz. Instead of probing the qubit itself, the team infers the noise from the photon loss it induces in a coupled superconducting cavity.&lt;/p&gt;
&lt;p&gt;The researchers validated the technique by injecting artificial noise into the system. The measured photon loss scaled in proportion to the noise they added. They argue the method reaches a higher-frequency spectral window than standard qubit-based spectroscopy. It could characterise noise while a quantum computation is actually running, though statistical limits still constrain its sensitivity at lower noise powers.&lt;/p&gt;
&lt;p&gt;Neither paper has yet been through peer review. Both papers’ authors still need to clear the same two hurdles: journal submission and independent replication by other labs.&lt;/p&gt;
</content:encoded><dc:creator>Owen Hartley</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-6_Transmons_without_Cheesing_001_1IbJXV.jpg" medium="image"/><category>computing</category></item><item><title>What Does QuTech in Delft Do, and What Has It Achieved?</title><link>https://explorequantum.com/article/what-does-qutech-in-delft-do-and-what-has-it-achieved/</link><guid isPermaLink="true">https://explorequantum.com/article/what-does-qutech-in-delft-do-and-what-has-it-achieved/</guid><description>QuTech is a Delft research institute working on a quantum computer and a quantum internet. Founded in 2015, it has built a three-node entangled quantum network and a seven-qubit logical qubit with error correction.</description><pubDate>Sun, 15 Nov 2026 10:21:51 GMT</pubDate><content:encoded>&lt;p&gt;According to TU Delft, QuTech is a Delft research institute with a dual mission. It aims to develop scalable prototypes of both a quantum computer and an inherently safe quantum internet, built on the laws of quantum mechanics. Its work is organised into three divisions: Quantum Computing, Quantum Internet and Qubit Research. Those divisions span two faculties of Delft University of Technology (TU Delft), Applied Sciences and Electrical Engineering, Mathematics and Computer Science.&lt;/p&gt;
&lt;h2 id=&quot;how-qutech-started&quot;&gt;How QuTech started&lt;/h2&gt;
&lt;p&gt;According to QuTech, TU Delft and the Netherlands Organisation for Applied Scientific Research (TNO) founded the institute in 2015. That followed a decision announced in October 2013, after earlier research by the Delft group of professor Leo Kouwenhoven. On 1 June 2015, TU Delft, TNO, the Dutch Research Council (NWO) and the Dutch ministries of Economic Affairs and of Education, Culture and Science signed a covenant. That covenant supports QuTech through 2025. A year earlier, in 2014, the Dutch government had, according to QuTech, already given its quantum research the status of “National Icon”. It cited the pioneering character of the work, its economic potential for the Netherlands, and its possible contribution to major societal challenges.&lt;/p&gt;
&lt;p&gt;QuTech says it now employs more than 300 people. It is based at Lorentzweg 1 in Delft, currently in the B and E wings of TU Delft’s Applied Sciences building. A dedicated new building is in development on the south side of the TU Delft campus. According to QuTech, it works within formal partnerships with TNO and NWO. It also runs collaboration programmes with the companies Intel and Fujitsu on quantum technology. Since January 2025, general director Kees Eijkel has led the institute, reporting to TU Delft’s executive board. Lieven Vandersypen is chief scientist.&lt;/p&gt;
&lt;h2 id=&quot;a-quantum-network-that-keeps-three-nodes-entangled-at-once&quot;&gt;A quantum network that keeps three nodes entangled at once&lt;/h2&gt;
&lt;p&gt;QuTech’s clearest result on the internet side, according to TU Delft, came on 15 April 2021. Its researchers reported in the journal Science that they had built the first multi-node quantum network based on entanglement. Three quantum processors, named Alice, Bob and Charlie, sat in the same building. The middle node, Bob, carried its own quantum memory and physical links to both of the other nodes. That let it maintain entanglement with Alice and Charlie at the same time. That is the basic building block a future quantum internet needs: a node that can hold onto one entangled link while a second one is set up. At the time, the team’s next step was to test the same approach outside the lab, on existing telecom fibre. It targeted a first metropolitan link for 2022.&lt;/p&gt;
&lt;h2 id=&quot;a-logical-qubit-built-from-seven-physical-qubits&quot;&gt;A logical qubit built from seven physical qubits&lt;/h2&gt;
&lt;p&gt;On the computing side, QuTech researchers announced in December 2021 that they had built a “logical qubit”. The result appeared in Nature Physics. The logical qubit was made of seven physical qubits, superconducting transmons. The team carried out all three basic operations on it, initialisation, a gate operation and readout, while errors were simultaneously detected and corrected. A logical qubit encodes one unit of quantum information redundantly across several physical qubits. That way, errors on individual qubits can be caught and fixed without destroying the information itself; it is the approach every large-scale quantum computer will need to control noise. After the seven-qubit result, the team aimed to scale the same approach up. Its next targets were 17 and later 49 physical qubits per logical qubit.&lt;/p&gt;
&lt;h2 id=&quot;where-it-stands&quot;&gt;Where it stands&lt;/h2&gt;
&lt;p&gt;QuTech pursues both halves of its mission in parallel. One track is a quantum computer built from error-corrected logical qubits; the other is a quantum internet built from entangled network nodes, with qubit research as a shared foundation. Its 2021 results on a three-node entangled network and a seven-qubit logical qubit remain the institute’s two clearest public milestones. Its partnerships with Intel and Fujitsu, under general director Kees Eijkel since January 2025, carry that work forward.&lt;/p&gt;
</content:encoded><dc:creator>Mara Quinlan</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-QuTech_Lab_Tour_(26819092810)_1h1IS7.jpg" medium="image"/><category>computing</category></item><item><title>What Is &quot;Harvest Now, Decrypt Later&quot;, and Which Data Is at Risk?</title><link>https://explorequantum.com/article/harvest-now-decrypt-later/</link><guid isPermaLink="true">https://explorequantum.com/article/harvest-now-decrypt-later/</guid><description>Harvest now, decrypt later is a surveillance strategy in which an adversary stores today&apos;s unreadable encrypted traffic and waits for a quantum computer able to break it; NIST, the UK&apos;s NCSC and ENISA have each set out how organisations should migrate before that day arrives.</description><pubDate>Sun, 15 Nov 2026 08:20:44 GMT</pubDate><content:encoded>&lt;p&gt;NIST’s final report IR 8547, published on 27 October 2026, deprecates every quantum-vulnerable public-key algorithm, including RSA and elliptic-curve cryptography, after 2030, according to NIST. NIST disallows their use under its own standards after 2035. That cut-off exists because adversaries are not waiting for a quantum computer to arrive before they start collecting; they are stockpiling encrypted traffic now, against the day one does.&lt;/p&gt;
&lt;h2 id=&quot;why-todays-encryption-is-tomorrows-problem&quot;&gt;Why today’s encryption is tomorrow’s problem&lt;/h2&gt;
&lt;p&gt;Security agencies call this strategy harvest now, decrypt later, according to Wikipedia. An adversary stores data it cannot read today, in the hope that a future quantum computer will let it decrypt that data years from now. It is also called store now decrypt later, steal now decrypt later or retrospective decryption. It only pays off if a powerful quantum computer able to run Shor’s algorithm is eventually built, according to Wikipedia. That algorithm can break the RSA and elliptic-curve encryption that protects most of today’s internet traffic. The hypothetical moment a quantum computer becomes powerful enough to run that attack is nicknamed “Q-Day”, with the related term “Y2Q” drawing a deliberate analogy to the Y2K date-change scare. Nobody has to wait for Q-Day to start harvesting. An adversary can intercept and archive encrypted traffic now, while it is still unreadable, and hold it until a capable machine exists.&lt;/p&gt;
&lt;p&gt;That makes the strategy most dangerous for data with a long confidentiality lifetime, meaning it has to stay secret for years or decades, according to Wikipedia. Named examples are diplomatic communications, personal health records, critical-infrastructure logs and intellectual property: worthless to an eavesdropper today, valuable once it can be decrypted. Data that only needs to stay confidential for days or weeks carries far less of this risk, since it will likely have lost its value long before any quantum computer arrives.&lt;/p&gt;
&lt;h2 id=&quot;a-warning-repeated-for-half-a-decade&quot;&gt;A warning repeated for half a decade&lt;/h2&gt;
&lt;p&gt;According to Wikipedia, the European Data Protection Supervisor wrote about the quantum threat to cryptography in 2020. The UK’s National Quantum Technologies Programme published “Quantum-Safe Secure Communications” in 2021. The US federal government proposed its migration roadmap in 2022, and CISA, the NSA and NIST jointly published quantum-readiness guidance in 2023. Germany’s BSI and the US Federal Reserve each published their own assessment of the risk in 2025.&lt;/p&gt;
&lt;h2 id=&quot;the-main-defence-and-a-narrower-one&quot;&gt;The main defence, and a narrower one&lt;/h2&gt;
&lt;p&gt;The main defence against harvest now, decrypt later is migrating encryption to post-quantum cryptography (PQC): algorithms believed to resist attack even from a quantum computer, according to Wikipedia. A narrower, complementary technique is physical-layer or optical-layer security. It alters the optical signal itself in fibre-optic links, so a standard receiver cannot record a usable copy of the transmission at all, denying the harvesting step rather than the later decryption.&lt;/p&gt;
&lt;p&gt;Three bodies have turned that defence into dated schedules. NIST’s is the one above: new use of RSA and elliptic-curve cryptography deprecated after 2030, disallowed under its own standards after 2035. The NCSC sets a three-phase timeline for UK organisations, according to the NCSC. By 2028, they should identify which cryptographic services need upgrading and build a migration plan. By 2031, they should have carried out their highest-priority upgrades. By 2035, migration to post-quantum cryptography should be complete across all their systems, services and products. ENISA, the EU’s cybersecurity agency, states that quantum computing will be highly disruptive for current security equipment and cryptography. It directs its mitigation work at integrating post-quantum cryptography with the systems organisations already run, rather than waiting to replace them outright.&lt;/p&gt;
&lt;h2 id=&quot;deadlines-side-by-side&quot;&gt;Deadlines side by side&lt;/h2&gt;
&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Body&lt;/th&gt;
&lt;th&gt;What it has set&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;NIST (United States)&lt;/td&gt;
&lt;td&gt;New use of RSA and ECC deprecated after 2030, disallowed under NIST standards after 2035.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;NCSC (United Kingdom)&lt;/td&gt;
&lt;td&gt;Migration plan by 2028, highest-priority upgrades by 2031, migration complete by 2035.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;ENISA (EU)&lt;/td&gt;
&lt;td&gt;No fixed date; focuses on integrating PQC with existing cryptographic systems.&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p&gt;Every one of these deadlines sits years before most experts expect Q-Day itself. That is why NIST, the NCSC and ENISA all set migration work in motion now, rather than waiting for that moment to arrive. Two earlier explainers on this site cover related ground: &lt;a href=&quot;/article/q-day-estimates/&quot;&gt;one on expert timelines&lt;/a&gt;, and &lt;a href=&quot;/article/pqc-deadlines-by-country/&quot;&gt;one on national migration deadlines&lt;/a&gt;.&lt;/p&gt;
</content:encoded><dc:creator>Owen Hartley</dc:creator><media:content url="https://explorequantum.com/_astro/photo-1614064641938-3bbee52942c7_ZlgzHM.jpg" medium="image"/><category>security</category></item><item><title>QuTech Links Quantum Inspire to DelftBlue Supercomputer for a Power Grid Test</title><link>https://explorequantum.com/article/quantum-inspire-delftblue-power-grid-test/</link><guid isPermaLink="true">https://explorequantum.com/article/quantum-inspire-delftblue-power-grid-test/</guid><description>TU Delft researcher Zeynab Kaseb ran the first workflow connecting Quantum Inspire to the DelftBlue supercomputer, testing a two-bus power grid optimisation on the Tuna-17 processor. The six-qubit version found the optimal answer in fewer runs but landed closer to the true result.</description><pubDate>Sun, 15 Nov 2026 02:10:29 GMT</pubDate><content:encoded>&lt;p&gt;QuTech has connected its Quantum Inspire cloud platform directly to DelftBlue, TU Delft’s classical supercomputer, according to QuTech. TU Delft researcher Zeynab Kaseb became the first person to run the combined workflow. The setup lets a user prepare and optimise calculations on DelftBlue. It then sends the resulting quantum circuits through Quantum Inspire to run on QuTech’s Tuna-17 processor, with results sent back for analysis. Kaseb works at TU Delft’s Faculty of Electrical Engineering, Mathematics and Computer Science, according to QuTech.&lt;/p&gt;
&lt;p&gt;She used the new link to tackle a power grid problem, according to QuTech. It was power-flow calculations for a two-bus electrical network, run with the Quantum Approximate Optimization Algorithm, or QAOA. Kaseb tested two circuit encodings of that problem on Tuna-17, each measured 1,024 times. The first used four qubits and 30 circuit layers. The second used six qubits and 165 layers.&lt;/p&gt;
&lt;p&gt;The two encodings landed on different trade-offs. The four-qubit run found the optimal solution in 215 of the 1,024 measurements, about 21 percent of the time, according to QuTech. The six-qubit run found it far less often, according to QuTech: just 16 of 1,024 measurements, about 1.56 percent. But the six-qubit version still came closer to the true answer when it missed the exact optimum. Its best candidate solution had a residual of 0.003262, against 0.01053 for the four-qubit version. More qubits and circuit layers meant a lower hit rate, but a tighter near-miss.&lt;/p&gt;
&lt;p&gt;Martin van Gijzen, scientific director of the Delft High Performance Computing Centre, said the test points to more such work ahead, according to QuTech. “We expect much of the integration of classical HPC computers and quantum computers,” he said. Kaseb’s test, he added, paves the way for more results combining the two kinds of machines.&lt;/p&gt;
&lt;p&gt;A two-bus network is a toy version of the power-flow problem grid operators actually face, and QuTech’s processor leaves little room to scale the encoding up much further. DelftBlue and Quantum Inspire now talk to each other directly: a user can move a problem from classical preprocessing to a quantum processor and back, inside one workflow. Whether that pipeline can carry a power-flow problem large enough for an actual grid operator is the next test ahead for Van Gijzen’s centre and Kaseb’s group.&lt;/p&gt;
</content:encoded><dc:creator>Owen Hartley</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-QuTech_Lab_Tour_(26819079000)_Z2vRi1t.jpg" medium="image"/><category>computing</category></item><item><title>JIJ, LTM and Kipu Quantum Add New Tools to IBM&apos;s Quantum Functions Catalog</title><link>https://explorequantum.com/article/jij-ltm-kipu-ibm-quantum-catalog/</link><guid isPermaLink="true">https://explorequantum.com/article/jij-ltm-kipu-ibm-quantum-catalog/</guid><description>Tokyo&apos;s JIJ, India&apos;s LTM and Kipu Quantum have each published a new function in IBM&apos;s Quantum Functions Catalog, covering optimisation, error mitigation and feature extraction for machine learning.</description><pubDate>Sat, 14 Nov 2026 20:09:29 GMT</pubDate><content:encoded>&lt;p&gt;Tokyo-based optimisation software developer JIJ has published its SQOA Solver in IBM’s Quantum Functions Catalog, according to IBM’s own listing for the tool. SQOA Solver maps up to three binary decision variables onto each physical qubit. It does that with a quantum random-access-code technique, combined with classical subspace diagonalisation. That lets users tackle combinatorial optimisation problems with more variables than the available qubits would normally allow. It also skips the repeated variational-parameter loops conventional methods need.&lt;/p&gt;
&lt;p&gt;LTM, part of India’s Larsen &amp;amp; Toubro Group, has added its AI-QEM error-mitigation software to the same catalogue, according to IBM’s Quantum Functions Catalog. LTM says that makes it the first global systems integrator to publish a function on the platform. AI-QEM combines a graph convolutional network with a transformer model to suppress hardware noise in quantum circuits. It needs no extra quantum-processor runtime, circuit repetitions or additional qubits. It is available to IBM Quantum Premium, Flex and on-premises plan users, running circuits of up to 100 qubits on IBM’s Heron r2 processors.&lt;/p&gt;
&lt;p&gt;A third catalogue addition comes from Kipu Quantum. It already offers its Iskay Quantum Optimizer for combinatorial optimisation in the same catalogue. Its new Rimay tool extracts features for machine learning. Rimay encodes tabular training data into a Hamiltonian, mapping each data feature to a qubit. It then runs counterdiabatic evolution on IBM quantum hardware, to produce quantum-derived features. Those features feed into classical machine-learning models as an added preprocessing step, rather than replacing those models. The tool is available to IBM Quantum Flex and Premium plan users.&lt;/p&gt;
&lt;p&gt;Kipu Quantum says it has already put Rimay to work with a Fortune 100 company. The goal there was to improve that company’s customer-intent prediction. Kipu Quantum calls it part of a broader push into quantum machine learning.&lt;/p&gt;
</content:encoded><dc:creator>Owen Hartley</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-IBM_Quantum_Computer_Demo_at_ITUWTSA_2024__Delhi_1_PgtXc.jpg" medium="image"/><category>business</category></item><item><title>How Can You Use a Quantum Computer Through the Cloud, and What Does It Cost?</title><link>https://explorequantum.com/article/how-to-use-a-quantum-computer-through-the-cloud/</link><guid isPermaLink="true">https://explorequantum.com/article/how-to-use-a-quantum-computer-through-the-cloud/</guid><description>Nobody needs to own a quantum computer to run one: IBM, Microsoft&apos;s Azure Quantum, Amazon Braket and D-Wave&apos;s Leap all sell cloud access to real hardware, with free tiers and per-minute or per-shot prices reported in November 2026.</description><pubDate>Sat, 14 Nov 2026 16:36:54 GMT</pubDate><content:encoded>&lt;p&gt;Nobody needs to buy, build or host a quantum computer to run a program on one. IBM, Microsoft’s Azure Quantum, Amazon Braket and D-Wave’s Leap all let a developer write code on an ordinary computer and submit it over the internet, according to the providers’ own documentation. Each platform then queues it for time on a real quantum processor housed at the provider’s own site and reads back the result, as reported in November 2026. A classical simulator of the hardware is usually available too, for testing before a real run.&lt;/p&gt;
&lt;h2 id=&quot;ibms-free-monthly-allowance-then-by-the-minute&quot;&gt;IBM’s free monthly allowance, then by the minute&lt;/h2&gt;
&lt;p&gt;IBM Quantum Platform gives anyone a free Open plan with up to 10 minutes of runtime a month, IBM says. Beyond that, Pay-As-You-Go billing starts at $96 per minute, charged per second with a one-second minimum. A Flex plan, paid upfront for at least 400 minutes a year, brings that down to $72 per minute. An annual Premium plan with dedicated support starts at $48 per minute for at least 5,200 minutes a year.&lt;/p&gt;
&lt;h2 id=&quot;azure-quantum-sells-the-door-but-partners-set-the-price&quot;&gt;Azure Quantum sells the door, but partners set the price&lt;/h2&gt;
&lt;p&gt;Microsoft’s Azure Quantum gives cloud access to hardware from several partner companies rather than a Microsoft-built quantum computer, and each partner sets and controls its own prices within the service, Microsoft says. A new Azure account separately gets a general $200 credit to explore Azure for 30 days, according to Microsoft. Microsoft’s own account FAQ says that credit cannot be applied to Azure Marketplace offers, the channel Azure Quantum’s partner hardware is billed through.&lt;/p&gt;
&lt;p&gt;The partners price very differently. IonQ bills in its own unit, the Azure Quantum Token, which combines a per-program minimum with a per-gate-shot rate, Microsoft’s Azure Quantum documentation says. Pay-As-You-Go access to its 25-qubit Aria 1 starts at a $12.42 minimum per run with error mitigation off, or $97.50 with it on. A monthly Aria-Forte subscription instead costs $25,000, plus Azure infrastructure costs. Pasqal charges EUR 3,000 per QPU-hour, or EUR 15 per emulator-hour, on its 100-qubit Fresnel processor, plus infrastructure costs, according to Microsoft’s Azure Quantum documentation. Rigetti charges $0.02 per 10-millisecond increment of job execution time on its Cepheus-1-108Q chip, with no separate per-shot fee. Quantinuum offers either Pay-As-You-Go on its own credit unit or a monthly subscription starting at $125,000.&lt;/p&gt;
&lt;h2 id=&quot;amazon-braket-charges-a-flat-fee-per-task-plus-a-fee-per-shot&quot;&gt;Amazon Braket charges a flat fee per task, plus a fee per shot&lt;/h2&gt;
&lt;p&gt;Amazon’s AWS offers its own route to several providers’ hardware through a service called Amazon Braket, which Amazon says connects users to processors from five hardware providers, AQT, IonQ, IQM, QuEra and Rigetti. A real quantum computer costs a flat $0.30 per task, plus a per-shot fee that varies by processor, Amazon says. That fee ranges from $0.000425 a shot on Rigetti’s Cepheus chip up to $0.08 a shot on IonQ’s Forte. Amazon’s AWS Free Tier covers only simulator time on Braket, one hour a month for an account’s first twelve months.&lt;/p&gt;
&lt;h2 id=&quot;d-wave-offers-a-trial-not-a-published-rate&quot;&gt;D-Wave offers a trial, not a published rate&lt;/h2&gt;
&lt;p&gt;According to D-Wave, it gives cloud access to its quantum annealers, a type of hardware built for optimisation problems rather than the general-purpose processors IBM, Azure and Braket offer, through a service called Leap. Leap’s own pricing page lists no self-serve free tier or published hourly rate, only a route to apply. D-Wave’s separate Leap Quantum LaunchPad program offers a 3-month trial, cost unstated, with real-time access to its quantum computers and project support. It is an exclusive trial, and D-Wave says not every applicant is accepted.&lt;/p&gt;
&lt;h2 id=&quot;where-to-start&quot;&gt;Where to start&lt;/h2&gt;
&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Provider&lt;/th&gt;
&lt;th&gt;Free access&lt;/th&gt;
&lt;th&gt;Paid access from&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;IBM Quantum Platform&lt;/td&gt;
&lt;td&gt;10 minutes/month, Open plan.&lt;/td&gt;
&lt;td&gt;$96/minute, Pay-As-You-Go.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Azure Quantum&lt;/td&gt;
&lt;td&gt;General $200 Azure credit, not valid for partner hardware.&lt;/td&gt;
&lt;td&gt;Varies by hardware partner.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Amazon Braket&lt;/td&gt;
&lt;td&gt;1 hour/month simulator, first 12 months.&lt;/td&gt;
&lt;td&gt;$0.30/task plus $0.000425–$0.08/shot.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;D-Wave Leap&lt;/td&gt;
&lt;td&gt;3-month LaunchPad trial, by application only.&lt;/td&gt;
&lt;td&gt;No published rate; apply for access.&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p&gt;For a first test run, IBM’s Open plan needs no purchase at all. Moving past it, on any of the four platforms, is a matter of minutes, tasks or shots rather than buying a machine.&lt;/p&gt;
</content:encoded><dc:creator>Owen Hartley</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-D-Wave_Two_installed_in_the_NASA_Advanced_Supercomputing_Division_1fVWJW.jpg" medium="image"/><category>business</category></item><item><title>Atom Computing&apos;s $100 Million Commerce Deal Carries a Possible $300 Million DARPA Follow-On</title><link>https://explorequantum.com/article/atom-computing-finalizes-100-million-commerce-deal/</link><guid isPermaLink="true">https://explorequantum.com/article/atom-computing-finalizes-100-million-commerce-deal/</guid><description>Atom Computing has finalised a $100 million Commerce Department R&amp;D deal and could draw $300 million more through a DARPA programme; Qunova says it matched a chemistry benchmark at 80 qubits for a fraction of its earlier cost.</description><pubDate>Sat, 14 Nov 2026 14:18:00 GMT</pubDate><content:encoded>&lt;p&gt;Atom Computing has finalised a $100 million research and development agreement with the United States Department of Commerce, according to Atom Computing. The deal funds work on fault-tolerant, utility-scale quantum computing systems. It follows an earlier Letter of Intent between the two sides. A fault-tolerant system keeps working correctly even as individual qubits make errors. It does this by spreading each unit of stored information across many physical qubits. Chief executive Ben Bloom said the money supports moving the company’s logical-qubit architecture into repeatable commercial production. He said it also expands engineering and manufacturing capacity.&lt;/p&gt;
&lt;p&gt;The agreement is one of nine such Letters of Intent the Department of Commerce announced in May 2026 under the CHIPS Act, according to NIST. Together they are worth $2.013 billion, spread across two quantum foundries and seven computing companies. The companies cover neutral-atom, silicon-spin, superconducting, photonic and trapped-ion hardware. Commerce takes a minority, non-controlling equity stake in each company in exchange for the funding, NIST said. Atom Computing’s neutral-atom approach holds individual atoms in place with laser light and uses them directly as qubits, one of several competing hardware designs the programme covers.&lt;/p&gt;
&lt;p&gt;The $100 million is not the end of Atom Computing’s government money, according to the company. It could draw up to $300 million more through Stage C of DARPA’s Quantum Benchmarking Initiative. Atom Computing is the only neutral-atom company to have advanced that far. Its neutral-atom platform has already passed 1,000 qubits, the first platform to do so, according to Atom Computing.&lt;/p&gt;
&lt;p&gt;Elsewhere in quantum computing this week, Qunova Computing said it matched a classical chemistry benchmark using 80 qubits, according to Qunova. The run used an IBM Quantum System Two processor paired with RIKEN’s Fugaku supercomputer. The South Korean company’s HI-VQE algorithm simulated a 4Fe-4S iron-sulfur cubane cluster, a molecule long used as a hard test case for chemistry calculations. Chief executive Kevin Rhee said it is the first quantum approach to match the classical benchmark at that scale, according to Qunova. The run landed a ground-state energy of -327.2482 Hartree, according to Qunova Computing. That is close enough to a classical benchmark of -327.2471 Hartree to count as chemically accurate. Qunova said the run needed about 2,518 Fugaku node-hours. That is roughly 300 times less computing cost than its earlier quantum-diagonalisation approach on the same cluster.&lt;/p&gt;
&lt;p&gt;DARPA’s Quantum Benchmarking Initiative moves companies through sequential stages. Each stage tests whether a hardware roadmap can be independently verified, rather than taken on a company’s word alone. Atom Computing’s advance to Stage C puts it ahead of every other neutral-atom entrant in that programme, according to Atom Computing. Chemical accuracy, the bar Qunova says it cleared, is the standard chemists use to judge a simulated energy against real molecular behaviour, useful for predicting reactions.&lt;/p&gt;
&lt;p&gt;Atom Computing has not said when it expects a decision on the DARPA Stage C funding. Nor has it given a firm date for when its fault-tolerant, utility-scale system will be ready for commercial customers to actually use.&lt;/p&gt;
</content:encoded><dc:creator>Owen Hartley</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-QuEra_Aquila_Magneto-Optical_Trap_2_ZK4sl5.jpg" medium="image"/><category>computing</category></item><item><title>What Do Quantum Computers And AI Have To Do With Each Other, And What Is Hype?</title><link>https://explorequantum.com/article/what-do-quantum-computers-and-ai-have-to-do-with-each-other/</link><guid isPermaLink="true">https://explorequantum.com/article/what-do-quantum-computers-and-ai-have-to-do-with-each-other/</guid><description>Quantum computing and AI overlap in three ways, says Wikipedia&apos;s quantum machine learning overview, and researchers it quotes call much of the field a hype-condensate; Google DeepMind, NVIDIA and Quantinuum point to narrower uses already working.</description><pubDate>Sat, 14 Nov 2026 10:26:25 GMT</pubDate><content:encoded>&lt;p&gt;Quantum computing and artificial intelligence overlap in three distinct ways, according to Wikipedia’s overview of quantum machine learning, current as of November 2026. One strand tries to use quantum computers to speed up machine-learning tasks that normally run on classical hardware. A second runs classical machine learning on data that quantum systems produce, such as reconstructing a quantum state or learning a system’s Hamiltonian. A third, hybrid approach splits a task between a classical computer and a quantum device. It hands the quantum device only the subroutines it can actually help with.&lt;/p&gt;
&lt;p&gt;That framing matters because the first strand draws most of the scepticism. Quantum computers speeding up AI itself is the use case researchers doubt most. Wikipedia’s summary says the field remains mostly theoretical after years of papers. It finds no proven case where a quantum method beats a classical one on a real machine-learning task. It quotes physicist Ryan Sweke as unconvinced any such task actually needs a quantum computer rather than a classical one. It also quotes researcher Jacob Biamonte warning that mixing machine learning with “quantum” catalyses a “hype-condensate”.&lt;/p&gt;
&lt;p&gt;According to the preprint, one of the clearest cases of that hype being cut down dates to July 2018. Ewin Tang, then an undergraduate, posted it that month. It showed that a quantum algorithm for recommendation systems, once seen as a strong candidate for exponential quantum speedup, owed its advantage to an unrealistic assumption about input data. That advantage was not down to quantum computing itself. Tang built a classical algorithm under matching assumptions that outputs a comparable recommendation only polynomially slower. The result was later extended to other proposed quantum machine-learning speedups, including one for principal component analysis.&lt;/p&gt;
&lt;p&gt;Real, working combinations of quantum and AI exist today. They sit in the second and third categories, not the first. In November 2024, Google DeepMind and Google Quantum AI said their AlphaQubit decoder identifies and corrects errors on a quantum processor. AlphaQubit is a neural network built on the transformer architecture used in large language models. It does this more accurately than the leading non-AI decoding methods. Google reported 6% fewer errors than tensor-network decoding and 30% fewer than correlated matching on its Sycamore processor. The model trained first on hundreds of millions of simulated error patterns, then on real error data from Sycamore. Google said it kept working accurately on simulated runs of up to 100,000 correction rounds, despite being trained on only up to 25.&lt;/p&gt;
&lt;p&gt;NVIDIA has gone after the hybrid approach directly. According to NVIDIA, in October 2025 it introduced NVQLink, an open interconnect architecture. NVQLink is meant to tightly couple GPU-based supercomputers with quantum processors. That includes running quantum error-correction control algorithms over high-throughput, low-latency links. NVIDIA said 17 quantum-processor builders, 5 quantum-controller builders and 9 US national laboratories took part in developing it, among them Los Alamos, Oak Ridge and Sandia. It said NVQLink integrates with NVIDIA’s CUDA-Q software, so developers can build hybrid applications spanning CPUs, GPUs and quantum processors.&lt;/p&gt;
&lt;p&gt;Quantinuum’s recent work shows a third way AI can assist: designing the quantum program itself, not running on quantum hardware. In July 2026, Quantinuum said it had worked with NVIDIA and an unnamed major pharmaceutical company on GenQAI. GenQAI is a generative AI framework fine-tuned on simulated quantum data. It generates quantum circuits for a given chemistry problem, rather than a person hand-designing them. Quantinuum validated the AI-generated circuits on its Helios trapped-ion quantum computer, for imipramine, an antidepressant used as a model compound. It said the approach points toward faster, more repeatable ways to turn a chemistry problem into a working quantum program.&lt;/p&gt;
&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Company&lt;/th&gt;
&lt;th&gt;What AI does&lt;/th&gt;
&lt;th&gt;Reported&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Google DeepMind / Google Quantum AI&lt;/td&gt;
&lt;td&gt;A transformer-based neural network decodes and corrects quantum hardware errors.&lt;/td&gt;
&lt;td&gt;November 2024.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;NVIDIA&lt;/td&gt;
&lt;td&gt;An interconnect lets GPU supercomputers run error-correction control algorithms for quantum processors.&lt;/td&gt;
&lt;td&gt;October 2025.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Quantinuum &amp;amp; NVIDIA&lt;/td&gt;
&lt;td&gt;A generative AI model designs quantum circuits for a chemistry problem.&lt;/td&gt;
&lt;td&gt;July 2026.&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p&gt;None of these three cases claims that a quantum computer makes an AI model run faster, the use case the field itself calls unproven. AlphaQubit and NVQLink put AI and GPUs to work on quantum hardware; GenQAI puts quantum hardware to work checking an AI-designed circuit.&lt;/p&gt;
</content:encoded><dc:creator>Owen Hartley</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-Dilution_Fridge_with_Quantum_Processors_Enclosed_008_8K_1JIYAN.jpg" medium="image"/><category>computing</category></item><item><title>D-Wave Tells SEC It Was Named a Gartner Market Shaper in Quantum Computing</title><link>https://explorequantum.com/article/dwave-8k-filing-gartner-market-shaper/</link><guid isPermaLink="true">https://explorequantum.com/article/dwave-8k-filing-gartner-market-shaper/</guid><description>D-Wave Quantum filed an SEC Form 8-K on 12 November 2026 to disclose a Gartner Market Shaper recognition, with CEO Alan Baratz calling it validation of the company&apos;s technology and execution. D-Wave describes itself as the only dual-platform annealing and gate-model quantum computing provider.</description><pubDate>Sat, 14 Nov 2026 10:24:25 GMT</pubDate><content:encoded>&lt;p&gt;D-Wave Quantum filed a Form 8-K with the SEC on 12 November 2026, according to D-Wave’s SEC filing. The filing furnishes a press release as Exhibit 99.1 rather than including it in the form’s body. That release announces a Gartner Market Shaper recognition. The filing was made under Item 7.01, which covers Regulation FD disclosure, and Item 9.01, which covers financial statements and exhibits.&lt;/p&gt;
&lt;p&gt;CEO Alan Baratz said, according to D-Wave, that the recognition validates the potential of the company’s technology and its ability to execute. He said D-Wave is building the systems, software and partnerships needed for an accelerated path to practical, scalable quantum computers.&lt;/p&gt;
&lt;p&gt;D-Wave describes itself as the only dual-platform quantum computing provider offering both annealing and gate-model systems, according to D-Wave. The company says it serves more than 100 commercial, government and research organisations. It reaches them through on-premises systems and its Leap quantum cloud service. Quantum annealing targets optimisation problems; gate-model systems are the more general-purpose architecture most other quantum computing companies pursue. D-Wave sells both.&lt;/p&gt;
&lt;p&gt;Regulation FD disclosure lets a public company share material information with the market. Furnishing a press release as a filing exhibit is a standard way to do that.&lt;/p&gt;
</content:encoded><dc:creator>Owen Hartley</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-D-Wave-Washington-1000Q_ZfmDdX.jpg" medium="image"/><category>business</category></item><item><title>IonQ Names Gary Merrill CFO, Creates Accounting Officer Role, Discloses Share Resale</title><link>https://explorequantum.com/article/ionq-cfo-change-merrill-share-resale/</link><guid isPermaLink="true">https://explorequantum.com/article/ionq-cfo-change-merrill-share-resale/</guid><description>IonQ says chief financial officer Inder Singh is stepping down and will be succeeded by former Commvault CFO Gary Merrill, as the quantum computing company also creates a new chief accounting officer post and discloses a prospectus supplement covering the resale of over 34 million shares.</description><pubDate>Sat, 14 Nov 2026 06:22:11 GMT</pubDate><content:encoded>&lt;p&gt;IonQ is replacing its chief financial officer. According to IonQ, Inder Singh is stepping down from all roles at the company, effective November 10, 2026. Gary Merrill, previously chief financial officer of Commvault Systems, will take over as CFO and treasurer on December 3, 2026.&lt;/p&gt;
&lt;p&gt;For the weeks in between, according to IonQ, chairman and chief executive Niccolo de Masi will serve as interim principal financial officer. That covers the gap between Singh’s exit and Merrill’s start date.&lt;/p&gt;
&lt;p&gt;IonQ set Merrill’s starting pay at a $525,000 base salary, with a bonus target equal to 100% of salary. It also granted him initial equity worth roughly $40 million, combining time-based, performance-based and make-whole restricted stock units.&lt;/p&gt;
&lt;p&gt;IonQ is also creating a new chief accounting officer position, according to the company. Thomas Austin, who previously held finance and accounting roles at Juniper Networks, will fill it from November 16, 2026.&lt;/p&gt;
&lt;p&gt;According to Niccolo de Masi, IonQ’s chief executive, Merrill and Austin bring public-company finance and accounting experience that will matter as IonQ scales. De Masi pointed in particular to Merrill’s cybersecurity background. He also thanked Singh for his contributions to the company, both before and since joining management.&lt;/p&gt;
&lt;p&gt;The same filing discloses a prospectus supplement, filed November 13, 2026, according to IonQ. It covers the potential resale by certain selling stockholders of 34,580,646 shares of IonQ common stock, registered under a shelf registration IonQ filed in February 2025.&lt;/p&gt;
&lt;p&gt;Austin’s appointment takes effect November 16, 2026. Merrill formally succeeds Singh as CFO three weeks later, on December 3, 2026.&lt;/p&gt;
</content:encoded><dc:creator>Owen Hartley</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-IonQ_headquarters_College_Park_MD_Z2t5Kg.jpg" medium="image"/><category>business</category></item><item><title>DLR Tests a Cheaper Error-Mitigation Protocol on a 100-Qubit IBM Processor</title><link>https://explorequantum.com/article/dlr-error-mitigation-bnzne-protocol/</link><guid isPermaLink="true">https://explorequantum.com/article/dlr-error-mitigation-bnzne-protocol/</guid><description>The German Aerospace Center built an error-mitigation protocol called bnZNE and ran it on IBM&apos;s 156-qubit ibm_fez processor, using 100 qubits. DLR reports up to 15% better fidelity than standard error mitigation, though gains vary by benchmark.</description><pubDate>Sat, 14 Nov 2026 02:09:02 GMT</pubDate><content:encoded>&lt;p&gt;Researchers at the German Aerospace Center (DLR) have built an extended error-mitigation protocol, called benchmarked-noise zero-noise extrapolation, or bnZNE, and tested it on a 100-qubit slice of IBM’s quantum hardware. Zero-noise extrapolation runs a circuit at amplified noise levels and extrapolates back to the zero-noise result. DLR’s version anchors that correction to verifiable benchmark circuits instead. Those circuits mirror the native-gate structure of the actual application circuit, rather than relying on the idealised noise-folding factors standard zero-noise extrapolation uses. DLR combined bnZNE with three other techniques: Pauli twirling, dynamical decoupling and Twirled Readout Error eXtinction, or TREX. Each benchmark circuit needs no more device runs than standard error mitigation already uses, according to DLR.&lt;/p&gt;
&lt;p&gt;According to DLR, the team behind bnZNE, Joseph Harris, Kevin Lively and Peter Schuhmacher, validated the protocol on IBM’s 156-qubit ibm_fez processor, a Heron r2 device. The experiments used 100 of those qubits. They ran Trotterized circuits with up to 20 layers and as many as 1,980 CZ gates, alongside smaller classical noisy simulations. Trotterized circuits break a continuous quantum evolution into discrete steps to simulate a physical system.&lt;/p&gt;
&lt;p&gt;DLR reports that bnZNE improved on standard quantum error mitigation by up to 15% in fidelity on these 100-qubit circuits. Fidelity measures how closely a quantum computer’s output matches the noise-free result a circuit should produce. The team measured single-site observables and two-site correlations along a qubit chain. The gains were not uniform: the paper’s own figures show an added bias-mitigation step sometimes raised the root-mean-square error on individual benchmarks. Even so, it improved accuracy on average.&lt;/p&gt;
&lt;p&gt;The work was funded by DLR’s Quantum Computing Initiative together with Germany’s Federal Ministry of Research, Technology and Space. The team published its results as an arXiv preprint, posted 10 March 2026. It comes with an accompanying GitHub repository for the error-mitigation code. The preprint has not yet been peer-reviewed. DLR has not stated further hardware targets for bnZNE.&lt;/p&gt;
</content:encoded><dc:creator>Mara Quinlan</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-IBM_System_Q_hRBzh.jpg" medium="image"/><category>computing</category></item><item><title>Rigetti Folds Quantum Elements&apos; Simulator Into Its Error-Correction Pipeline</title><link>https://explorequantum.com/article/rigetti-folds-constellation-into-error-correction-pipeline/</link><guid isPermaLink="true">https://explorequantum.com/article/rigetti-folds-constellation-into-error-correction-pipeline/</guid><description>Rigetti has added Quantum Elements&apos; Constellation simulation platform to its error-correction workflow, after checking its predictions against results from the company&apos;s 36-qubit and 108-qubit processors.</description><pubDate>Fri, 13 Nov 2026 22:21:25 GMT</pubDate><content:encoded>&lt;p&gt;Rigetti Computing has, according to Rigetti and Quantum Elements, written Quantum Elements’ Constellation simulation platform into its error-correction pipeline. The tool gives Rigetti’s engineers a software model of its superconducting chips to test against before they book time on the physical hardware. The move extends a multi-year technical collaboration between the two companies.&lt;/p&gt;
&lt;p&gt;Constellation, according to Rigetti and Quantum Elements, builds a software model of a quantum chip. It runs physics simulations on ordinary computers, so engineers can try out circuit layouts, pulse timing and error-correction schemes. That spares them the scarce, costly time slots on the real processor. Rigetti and Quantum Elements say they checked those predictions against actual measurements from Rigetti’s 36-qubit and 108-qubit processors. They compared forecast and measured single- and two-qubit gate performance on both devices. Matching the model against two chip generations, not one, is what the companies point to as evidence it holds up as device size grows.&lt;/p&gt;
&lt;p&gt;Rigetti CEO Subodh Kulkarni says the tool has widened the company’s ability to model how gates behave on larger, multi-chip systems. He said it lets Rigetti study the error channels that limit performance, and that it now feeds directly into the company’s error-correction work. Quantum Elements CEO Izhar Medalsy, for his part, said he welcomed seeing Rigetti fold Constellation into its own engineering process to model how its hardware behaves.&lt;/p&gt;
&lt;p&gt;Both companies frame the step, according to Rigetti and Quantum Elements, as part of the longer push toward fault-tolerant quantum computing. Constellation’s simulated noise data runs through Rigetti’s error-correction pipeline, so engineers can estimate logical error rates and test design changes before loading them onto the chip. That sequencing matters, because error-correction work needs long runs of repeated measurements, exactly the kind of job that competes hardest for scarce processor time.&lt;/p&gt;
&lt;p&gt;The next step for Rigetti is to carry that simulator-guided error-correction work from the modelling stage into results run on its own chips.&lt;/p&gt;
</content:encoded><dc:creator>Mara Quinlan</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-TechCrunch_Disrupt_2018_-_Day_3_-_021_(43627534315)_eUP2S.jpg" medium="image"/><category>computing</category></item><item><title>Q-CTRL Raises $290 Million Series C Led by Salesforce Ventures</title><link>https://explorequantum.com/article/q-ctrl-series-c-funding-salesforce-ventures/</link><guid isPermaLink="true">https://explorequantum.com/article/q-ctrl-series-c-funding-salesforce-ventures/</guid><description>Australian quantum infrastructure software company Q-CTRL has closed a $290 million Series C round led by Salesforce Ventures, pushing its valuation above $1 billion and its disclosed funding past $400 million.</description><pubDate>Fri, 13 Nov 2026 18:24:02 GMT</pubDate><content:encoded>&lt;p&gt;Q-CTRL’s software helps quantum computers and sensors run more reliably by correcting errors in their hardware. According to Q-CTRL, the Australian company has now closed a $290 million Series C round led by Salesforce Ventures, the venture capital arm of US software maker Salesforce. The round, worth A$415 million, values Q-CTRL above $1 billion, making it a unicorn. Q-CTRL says it is the largest fundraise yet for a private quantum software firm.&lt;/p&gt;
&lt;p&gt;Q-CTRL says the round drew new backers including Affinity Partners, AMD Ventures, EDGE Group, Australia’s National Reconstruction Fund Corporation, MPower Partners and SPARX Asset Management. Returning investors included Lockheed Martin Ventures, In-Q-Tel, DCVC, Main Sequence, Square Peg and Bullhound Capital. The company describes the round as oversubscribed, saying it was expanded twice before it closed.&lt;/p&gt;
&lt;p&gt;Q-CTRL plans to put the new capital toward expanding its AI-driven quantum control software, the company says. It also wants to scale commercial deployments for customers in finance, energy, defense, autonomy and aerospace. On top of that, it intends to grow its reach beyond its current base in Australia and the United States.&lt;/p&gt;
&lt;p&gt;Q-CTRL was founded in 2017, according to the company. It is led by founder and chief executive Michael Biercuk. It is headquartered in Sydney, with offices in Los Angeles, San Francisco, Huntsville, Berlin and Oxford. Outlets report a workforce of 101 to 500 people. Its product line centres on the Fire Opal and Boulder Opal tools for error suppression and calibration. Fire Opal’s Dynamics Simulator is also distributed through IBM’s Quantum Functions Catalog, making it available to developers running circuits on IBM’s hardware.&lt;/p&gt;
&lt;p&gt;Q-CTRL says a third product, Ironstone Opal, is meant to keep aircraft, ships and drones navigating accurately without a GPS signal, using onboard quantum sensors instead of satellite signals that can be jammed or spoofed. It has become the first quantum navigation system to reach safety-of-flight qualification under the RTCA DO-160 airworthiness standard, Q-CTRL says. The company says it validated the system in flight to an accuracy of 0.3 nautical miles. That accuracy is about 100 times better than the best available alternative on the market today, Q-CTRL says. Q-CTRL has engagements on the system with commercial-aircraft maker Airbus and defense prime Lockheed Martin. It also offers a sub-1kg version aimed at small surveillance drones.&lt;/p&gt;
&lt;p&gt;The Series C builds on a Series B that, according to Q-CTRL’s own announcements, reached a $113 million total by October 2024. That extension was branded a B-2 round and led by GP Bullhound. It already folded in a $27.4 million Series B-1 round the company closed in January 2023. That B-1 round had brought the Series B total to $52.4 million at the time. With the new $290 million round added, Q-CTRL’s disclosed funding to date totals more than $400 million.&lt;/p&gt;
&lt;p&gt;That capital is now earmarked for scaling the deployments Q-CTRL already runs with customers in finance, energy, defense, autonomy and aerospace. The company also wants to push its reach into markets outside its current base in Australia and the United States.&lt;/p&gt;
</content:encoded><dc:creator>Owen Hartley</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-Sydney_Harbour_Bridge_night_Z2ibJpC.jpg" medium="image"/><category>business</category></item><item><title>What Can A Quantum Computer Do For Chemistry, Materials And Drug Research?</title><link>https://explorequantum.com/article/what-can-quantum-computers-do-for-chemistry/</link><guid isPermaLink="true">https://explorequantum.com/article/what-can-quantum-computers-do-for-chemistry/</guid><description>Simulating a molecule&apos;s electrons is often too hard for classical computers, says Google, which wants quantum computers to do that directly; Google, IBM, Pasqal and Quantinuum have used quantum hardware to model molecules, a far larger protein, and a drug compound.</description><pubDate>Fri, 13 Nov 2026 16:28:40 GMT</pubDate><content:encoded>&lt;p&gt;Simulating how electrons interact inside a molecule is often classically intractable, says Google, because of the complex quantum mechanics at play. A major ambition of quantum computers, Google says, is to simulate that same quantum mechanics directly. That idea has since moved into concrete runs. Labs and companies have used quantum hardware to model small molecules, a far larger protein, and a drug-related compound.&lt;/p&gt;
&lt;p&gt;Google frames the opportunity broadly. It calls the quantum approach a potentially invaluable new tool, competing against classical chemistry methods refined for more than eighty years. Google names battery cathode degradation, reactions with heterogeneous catalysts, and the electrical properties of graphene and superconductors as materials problems it says the approach could eventually help with.&lt;/p&gt;
&lt;p&gt;The first big proof point came from Google. According to Google, in August 2020 it reported in the journal Science that its 54-qubit Sycamore processor modelled linear chains of hydrogen atoms using 10 qubits. The run used a noise-robust variational quantum eigensolver. Google called it a calculation twice as large as previous chemistry work on a quantum computer, with ten times more quantum gate operations than that earlier work. It said the result showed its error-mitigation method reached the precision chemists need for real predictions.&lt;/p&gt;
&lt;p&gt;IBM took on a harder target with Lockheed Martin. In May 2025, the two reported using 52 qubits of an IBM processor to simulate methylene, a three-atom molecule with two unpaired electrons. The run used up to 3,000 two-qubit gates on a molecule that is hard for classical methods to model accurately. IBM said its sample-based quantum diagonalization method matched classical reference calculations and experiment on the molecule’s singlet-triplet energy gap. It called this the first time the technique had been applied to this kind of open-shell molecule, and linked the result to aerospace, combustion chemistry and sensor-design uses.&lt;/p&gt;
&lt;p&gt;A year later, IBM moved from a handful of atoms to a real protein. In May 2026, IBM, Cleveland Clinic and RIKEN said they had modelled trypsin, a 12,635-atom protein, with quantum hardware. The run used up to 94 qubits of IBM’s 156-qubit Quantum Heron processors and nearly 6,000 quantum operations in parts of the workflow. That protein is roughly 40 times larger than what the same method had handled six months earlier. IBM said accuracy at a key step of the workflow improved up to 210 times over that same period. It said modelling how a drug candidate binds to a protein target earlier and more accurately could shorten drug-development timelines that now often take a decade.&lt;/p&gt;
&lt;p&gt;Drug discovery is also where Europe’s Pasqal has focused its own work. In September 2023, Pasqal said it had run a hybrid quantum-classical algorithm with drug-discovery company Qubit Pharmaceuticals. The algorithm ran on its Fresnel 1 neutral-atom quantum computer and places water molecules inside protein pockets, including buried ones that are hard to model. Pasqal called it the first time a quantum algorithm had been used for a molecular-biology task of this kind. It said accurate water placement helps predict how a drug candidate can block a target protein’s harmful activity.&lt;/p&gt;
&lt;p&gt;The newest step tackles a different bottleneck: designing the quantum program itself. In July 2026, Quantinuum said it had worked with NVIDIA and an unnamed major pharmaceutical company on GenQAI, a generative AI framework for simulating pharmaceutical compounds. In it, an AI model fine-tuned on simulated quantum data generated the quantum circuits. Quantinuum validated those AI-generated circuits by running them on its Helios trapped-ion quantum computer for imipramine, an antidepressant chosen as a model compound for drug degradation and shelf-life studies. It said the work points toward more repeatable, scalable workflows for translating chemistry problems into quantum programs.&lt;/p&gt;
&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Who&lt;/th&gt;
&lt;th&gt;Molecule or target&lt;/th&gt;
&lt;th&gt;Hardware used&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Google&lt;/td&gt;
&lt;td&gt;Hydrogen chains H8, H10 and H12.&lt;/td&gt;
&lt;td&gt;54-qubit Sycamore processor, 10 qubits used, August 2020.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Pasqal &amp;amp; Qubit Pharmaceuticals&lt;/td&gt;
&lt;td&gt;Water placement in protein pockets.&lt;/td&gt;
&lt;td&gt;Fresnel 1 neutral-atom quantum computer, September 2023.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;IBM &amp;amp; Lockheed Martin&lt;/td&gt;
&lt;td&gt;Methylene (CH2).&lt;/td&gt;
&lt;td&gt;52 qubits, up to 3,000 two-qubit gates, May 2025.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;IBM, Cleveland Clinic &amp;amp; RIKEN&lt;/td&gt;
&lt;td&gt;Trypsin, a 12,635-atom protein.&lt;/td&gt;
&lt;td&gt;Up to 94 of 156 qubits on a Quantum Heron processor, May 2026.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Quantinuum &amp;amp; NVIDIA&lt;/td&gt;
&lt;td&gt;Imipramine, a model drug compound.&lt;/td&gt;
&lt;td&gt;Helios trapped-ion computer, AI-generated circuits, July 2026.&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
</content:encoded><dc:creator>Owen Hartley</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-IBM_Q_Hummingbird_quantum_processor_ZvxfTL.jpg" medium="image"/><category>computing</category></item><item><title>Two Japanese-Led Teams Push Quantum Sensing Into Crystals</title><link>https://explorequantum.com/article/japan-quantum-sensing-crystals-mof-trapped-ion/</link><guid isPermaLink="true">https://explorequantum.com/article/japan-quantum-sensing-crystals-mof-trapped-ion/</guid><description>A University of Tokyo and Northwestern team built a porous crystal forming spin-polarized radical pairs sensitive to surrounding chemistry. A University of Osaka team tracked a single phonon through a ten-ion crystal, a setup it says could scale toward multi-particle simulation.</description><pubDate>Fri, 13 Nov 2026 14:15:49 GMT</pubDate><content:encoded>&lt;p&gt;A team led by Nobuhiro Yanai at the University of Tokyo and Michael R. Wasielewski at Northwestern University built a porous metal-organic framework, or MOF, according to Nature Communications. Kyushu University collaborators joined the work. The crystal packs a light-absorbing donor molecule and an acceptor molecule into dense, well-defined pairs. That packing selectively produces strongly coupled, spin-polarized triplet radical pairs: pairs of unpaired electrons whose spin states come out aligned rather than random. A metal-organic framework is a crystal built from metal clusters linked by organic molecules into a regular, porous lattice. That structure fixes the donor and acceptor in a repeating arrangement, rather than letting them tumble freely in solution.&lt;/p&gt;
&lt;p&gt;According to Nature Communications, the dense packing efficiently quenches the porphyrin’s light-excited state and funnels it into those triplet radical pairs. The pairs show narrow electron paramagnetic resonance signals and long spin-lattice relaxation times. Those relaxation times persist across different guest molecules held in the crystal’s pores, but they also vary with which guest is present. That guest-dependence is what turns the effect from a laboratory curiosity into a sensing mechanism: the radical pair’s spin state reports on the chemistry around it. It extends a phenomenon previously studied in isolated molecules to a structurally organized material. The result is not yet a working sensor. It establishes that the chemical sensitivity survives inside a solid, porous crystal, which is the condition a deployable device would need. The paper on sublevel-selective electron spin hyperpolarization of triplet radical pairs in metal-organic frameworks appeared in Nature Communications on 13 November 2026.&lt;/p&gt;
&lt;p&gt;According to the arXiv preprint, a separate, unrelated team at the University of Osaka worked with a different kind of crystal entirely. Takumi Yuri, Yu-Xuan Chen and Kenji Toyoda built a one-dimensional chain of ten trapped calcium-40 ions, held in a harmonic trap that spaces them unevenly, wider at the edges and tighter near the centre. Using laser pulses, the team created a single local phonon, a discrete quantum of vibrational energy, and tracked it as it moved through the chain. The team describes this as one of the largest-scale demonstrations to date of single-quantum control over trapped-ion vibrations.&lt;/p&gt;
&lt;p&gt;The arXiv preprint reports that the phonon spread through the chain partly by hopping between neighbouring ions and partly through broader collective motion. The hopping rate between adjacent ions scales approximately as the inverse cube of the distance between them. That measured scaling matched the team’s numerical simulations down to fine detail. It also showed the phonon’s movement is highly sensitive to small changes in ion spacing. The arXiv preprint argues the setup could extend to many-body systems of bosonic particles and more complex quasiparticles. It discusses the Jaynes-Cummings-Hubbard model, which couples each ion’s internal state to its local vibrational mode, as a system the platform could help simulate. The arXiv preprint on quantum propagation of a vibrational excitation in a harmonically confined trapped-ion crystal was posted on 10 November 2026 and has not yet been peer-reviewed.&lt;/p&gt;
</content:encoded><dc:creator>Mara Quinlan</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-Yasuda_Auditorium_-_Tokyo_University_3_24pgyt.jpg" medium="image"/><category>sensing</category></item><item><title>Why Do Quantum Computers Need To Be Cooled To Near Absolute Zero, And Which Ones Don&apos;t?</title><link>https://explorequantum.com/article/why-quantum-computers-are-cold/</link><guid isPermaLink="true">https://explorequantum.com/article/why-quantum-computers-are-cold/</guid><description>Superconducting qubits, used by IBM and others, only work once cooled to about a hundred times colder than one degree above absolute zero, IBM says. Trapped-ion hardware from IonQ and neutral-atom hardware from QuEra skip that cryogenic plant and run at room temperature instead.</description><pubDate>Fri, 13 Nov 2026 10:47:27 GMT</pubDate><content:encoded>&lt;p&gt;Superconducting qubits only work once chilled to about a hundred times colder than one degree above absolute zero, achieved with supercooled superfluids, according to IBM. IBM, Google and Rigetti build their chips on this qubit type. Not every quantum computer needs that cold. Trapped-ion machines from IonQ and neutral-atom machines from QuEra keep their surrounding hardware at room temperature instead.&lt;/p&gt;
&lt;p&gt;According to IBM, the reason is decoherence: a qubit’s fragile quantum state collapsing into an ordinary, classical one. Any stray heat or electrical interference can trigger it. So IBM says large machines need cooling colder than outer space. IBM reaches that temperature with a dilution refrigerator, a device that mixes the helium isotopes helium-3 and helium-4. Vacuum pumps circulate and dilute helium-3 into the mixture until the target temperature is reached. Below about 870 millikelvin that mixture separates into two phases, and helium-3 atoms crossing into the dilute phase absorb heat from their surroundings, which produces the cooling effect, according to Wikipedia. IBM’s Goldeneye concept system, built to test this cooling approach at a larger scale, reached about 25 millikelvin with a quantum processor wired inside it. Below the operating threshold, the qubits both superconduct, carrying current with no resistance, and show quantum behaviour at all, according to IBM.&lt;/p&gt;
&lt;p&gt;Trapped-ion hardware cools only the qubits, not the machine around them. IonQ says it holds individual ions in a chip-based ion trap built from about 100 electrodes. Lasers then cool them, using a technique called Doppler cooling, to half of one one-thousandth of a degree above absolute zero. The rest of the hardware stays at room temperature, so no dilution refrigerator is needed. IonQ instead isolates the ion trap chip inside an ultra-high vacuum chamber, pumped down to around 10 to the power of minus 11 Torr. That is about a hundred trillion times fewer molecules per cubic inch than the air a person breathes, and it stops stray atoms from disturbing the ions’ quantum state.&lt;/p&gt;
&lt;p&gt;QuEra’s neutral-atom systems take a similar route. According to QuEra, individual rubidium atoms sit in an ultra-high vacuum cell. Lasers cool them optically to microkelvin temperatures, colder than any refrigerator reaches, with no cryogenic plant around them. That design choice shows up in power use. QuEra says its room-temperature Aquila system draws under 12 kilowatt-hours. Its coming Libra system is designed to use less than 40 kilowatts. Either figure is low enough for the machine to sit inside an existing data centre or HPC facility, without specialised cryogenic infrastructure.&lt;/p&gt;
&lt;p&gt;So the cooling a quantum computer needs depends on where its qubits come from. Superconducting qubits are circuits etched into a chip, so the whole chip must be cold enough to superconduct and hold a quantum state. Trapped-ion and neutral-atom qubits are individual atoms or ions, cooled one by one with lasers inside a vacuum. That leaves the surrounding electronics free to run at room temperature. As quantum computers move into data centres, that difference comes down to a dilution refrigerator on one side and a vacuum chamber with a laser system on the other.&lt;/p&gt;
</content:encoded><dc:creator>Mara Quinlan</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-XLD_Pulse-Tube-Type_Large_Frame_Dilution_Refrigerator_Cryostat_5UJOh.jpg" medium="image"/><category>computing</category></item><item><title>Warsaw Team&apos;s Rydberg Sensor Captures 3.3 Times More Microwave Information</title><link>https://explorequantum.com/article/warsaw-rydberg-microwave-sensor-fisher-information-gain/</link><guid isPermaLink="true">https://explorequantum.com/article/warsaw-rydberg-microwave-sensor-fisher-information-gain/</guid><description>University of Warsaw physicists built a Rydberg-atom microwave sensor that captures about 3.3 times more Fisher information than a non-interacting design, published in Nature Communications. Separately, Q-CTRL&apos;s simulator ran a many-body simulation 12 times faster on IBM&apos;s Nighthawk processor.</description><pubDate>Fri, 13 Nov 2026 10:35:58 GMT</pubDate><content:encoded>&lt;p&gt;According to the University of Warsaw team, a microwave sensor built from cold, interacting Rydberg atoms captured about 3.3 times more Fisher information than a comparable sensor using non-interacting atoms. Fisher information measures how precisely a signal’s strength can be estimated from a set of measurements. More of it means a sharper reading of the same microwave field. The University of Warsaw team published the result in Nature Communications. It was led by Stanisław Kurzyna and Bartosz Niewelt, under professors Michał Parniak and Wojciech Wasilewski.&lt;/p&gt;
&lt;p&gt;The Warsaw team traces the gain to interactions it deliberately engineered between the Rydberg atoms. That way, losses occurring during detection destroy less of the information the atoms already picked up from the microwave field. Rydberg atoms are electrons pushed into a very high energy state, which makes them highly sensitive to electric and microwave fields. Ordinarily, a chunk of that sensitivity is lost the moment the atoms are read out. The Warsaw group’s interaction scheme is built specifically to resist that loss.&lt;/p&gt;
&lt;p&gt;Separately, quantum control company Q-CTRL says its Fire Opal Dynamics Simulator ran a large fermionic many-body simulation 12 times faster on an IBM Quantum Nighthawk r2 processor than on an earlier IBM Heron r3 system. The run dropped from roughly four minutes 25 seconds to about 23 seconds. Its deviation from the expected answer stayed below 1%. Q-CTRL has now listed the simulator in IBM’s Quantum Functions Catalog. It has also handled a 60-site Fermi-Hubbard chain on 120 qubits, beyond the reach of exact classical simulation. In separate benchmark runs it reached circuits with up to 90 Trotter steps and close to 13,000 two-qubit gates. Q-CTRL reports that the simulator beat classical comparison methods on the same problem by wide margins. It ran more than 3,000 times faster than a CPU-based tensor-network baseline, and more than 400 times faster than GPU-accelerated classical solvers.&lt;/p&gt;
&lt;p&gt;A cold-atom sensor reading a microwave field and a superconducting processor simulating interacting particles are different machines, but both results come from squeezing more signal out of the same noisy hardware. Q-CTRL’s simulator is listed in IBM’s Quantum Functions Catalog, open to any user who wants to run it.&lt;/p&gt;
</content:encoded><dc:creator>Mara Quinlan</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-Wydzial_Fizyki_UW_Smyczkowa-Faculty_of_Physics_of_UW_Smyczkowa_street_ZzOJqP.jpg" medium="image"/><category>computing</category></item><item><title>Pasqal Formalises AtomIQ-Lab, a Joint Lab With CNRS on Neutral-Atom Qubits</title><link>https://explorequantum.com/article/pasqal-cnrs-atomiq-lab/</link><guid isPermaLink="true">https://explorequantum.com/article/pasqal-cnrs-atomiq-lab/</guid><description>Pasqal has formalised AtomIQ-Lab, a joint laboratory with CNRS and the Institut d&apos;Optique Graduate School in France, targeting two physics problems it says stand between today&apos;s neutral-atom processors and larger, error-corrected machines.</description><pubDate>Fri, 13 Nov 2026 02:10:04 GMT</pubDate><content:encoded>&lt;p&gt;Pasqal has formalised AtomIQ-Lab, a joint laboratory with CNRS and the Institut d’Optique Graduate School, according to Pasqal. The lab was inaugurated in November 2025. It targets physics challenges in scaling neutral-atom qubits and is meant to speed up moving research results into working technology.&lt;/p&gt;
&lt;p&gt;AtomIQ-Lab’s stated priorities are two physics problems, according to Pasqal. One is lossless measurement of atomic states, needed for fault-tolerant operation. The other is continuous reloading of atom registers during operation. Pasqal says both stand between today’s neutral-atom processors and larger, error-corrected machines.&lt;/p&gt;
&lt;p&gt;Its neutral-atom technology grew out of experiments at Laboratoire Charles Fabry, a joint CNRS, Institut d’Optique and Université Paris-Saclay research unit, according to Pasqal. Co-founders Antoine Browaeys and Thierry Lahaye worked there before starting Pasqal in 2019. The pair now serve as scientific advisors to the company. The two sides have run six joint CIFRE PhD theses since then, four of them still ongoing.&lt;/p&gt;
&lt;p&gt;Pasqal set up a dedicated research chair at the Institut d’Optique in October 2025 for a physicist working on topics the lab shares with the company. That followed an earlier Institut d’Optique-Pasqal workshop in 2025 that brought together 50 to 60 researchers from both sides to identify the priority areas AtomIQ-Lab now works on.&lt;/p&gt;
&lt;p&gt;No independent outlet has yet reported on AtomIQ-Lab’s staffing, funding or site details beyond the research chair and the CIFRE theses.&lt;/p&gt;
</content:encoded><dc:creator>Mara Quinlan</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-Institut_d&apos;Optique_Graduate_School_-_Palaiseau_6Appu.jpg" medium="image"/><category>computing</category></item><item><title>IBM Unveils 520-Qubit Nighthawk r3 Chip, Charts Two-Chip Route to 1,040 Qubits</title><link>https://explorequantum.com/article/ibm-nighthawk-r3-520-qubit/</link><guid isPermaLink="true">https://explorequantum.com/article/ibm-nighthawk-r3-520-qubit/</guid><description>IBM has launched Nighthawk r3, a 520-qubit processor it says can run circuits of up to 10,000 accurate two-qubit operations, and set out a plan to link two such chips into a system of more than 1,040 qubits in 2027.</description><pubDate>Thu, 12 Nov 2026 22:13:53 GMT</pubDate><content:encoded>&lt;p&gt;IBM has unveiled Nighthawk r3, a superconducting processor with 520 programmable qubits built from 2,034 physical qubits, according to the company. Programmable qubits are the ones a user’s circuit can address directly. The larger physical count includes qubits used for wiring and control. IBM announced the chip at its Quantum Developer Conference (QDC 2026) in Yorktown Heights, New York. It succeeds the Nighthawk r2 chip IBM introduced in September 2026.&lt;/p&gt;
&lt;p&gt;IBM says Nighthawk r3 can run circuits of up to 10,000 accurate two-qubit operations, up from more than 7,500 on Nighthawk r2. A two-qubit operation is a step that entangles or otherwise combines two qubits. More of them running accurately in a row lets a circuit tackle harder problems before noise overwhelms the result.&lt;/p&gt;
&lt;p&gt;IBM plans to link two Nighthawk r3 chips into a single system of more than 1,040 qubits in 2027. It says it has already connected Nighthawk r1 processors with four so-called l-couplers, hardware links between chips. Those links reached 99.3 percent quantum-state transfer fidelity, and IBM is working toward a 99.9 percent target for the chip-to-chip links the 1,040-qubit system will need. That linked system is a step toward IBM’s longer-term IBM Quantum Starling machine, targeted for 2029.&lt;/p&gt;
&lt;p&gt;IBM says clients and partners, including E.ON, Algorithmiq and Q-CTRL, have run more than 34 billion circuits on Nighthawk r2 since it was deployed. Among the results IBM cites: an E.ON quantum-machine-learning workload on its energy grid ran 15 times faster. A magnetic-materials simulation run with Oak Ridge National Laboratory ran 12 times faster, and an Algorithmiq quantum-advantage experiment ran 8 times faster.&lt;/p&gt;
&lt;p&gt;Jay Gambetta, Director of IBM Research and an IBM Fellow, said useful quantum computing depends on delivering results with both accuracy and speed. As IBM scales the Nighthawk architecture to larger machines, he said, it is integrating more closely with classical computing resources on the way to fault tolerance. Fault tolerance is the point at which a quantum computer corrects its own errors faster than they build up.&lt;/p&gt;
&lt;p&gt;Alongside the chip, IBM says it has brought high-performance classical and quantum computing together into a single managed environment on the IBM Quantum Platform. That lets users connect quantum workflows directly to classical compute, from laptops to supercomputers. IBM names error correction, error mitigation, noise learning and post-processing as tasks it covers.&lt;/p&gt;
&lt;p&gt;The next marker to watch is the 1,040-qubit linked system IBM is targeting for 2027, on its way to the Starling machine planned for 2029.&lt;/p&gt;
</content:encoded><dc:creator>Mara Quinlan</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-IBM_Quantum_System_One_ZpXCss.jpg" medium="image"/><category>computing</category></item><item><title>Qunova Says Its Hybrid Algorithm Matched a Classical Benchmark on an Iron-Sulfur Cluster</title><link>https://explorequantum.com/article/qunova-hi-vqe-iron-sulfur-benchmark/</link><guid isPermaLink="true">https://explorequantum.com/article/qunova-hi-vqe-iron-sulfur-benchmark/</guid><description>South Korea&apos;s Qunova Computing says its HI-VQE algorithm reached chemical accuracy on a notoriously hard iron-sulfur molecule using 80 qubits, a result the company says beats its own 72-qubit run from September at a fraction of the computing cost.</description><pubDate>Thu, 12 Nov 2026 18:43:20 GMT</pubDate><content:encoded>&lt;p&gt;South Korea’s Qunova Computing says its HI-VQE algorithm has reached chemical accuracy on a 4Fe-4S iron-sulfur cluster. The run used 80 qubits on an IBM Quantum System Two processor, according to Qunova. The company calls the cluster one of the hardest known test cases in computational chemistry. HI-VQE pairs that IBM processor with RIKEN’s Fugaku supercomputer in a hybrid workflow, according to Qunova.&lt;/p&gt;
&lt;p&gt;Qunova’s estimated ground-state energy for the cluster is about -327.2482 Hartree. That falls within the chemical-accuracy range of the classical DMRG reference value of about -327.2471 Hartree the company used as its benchmark. Qunova’s chief executive, Kevin Rhee, said the cluster has long been a test case only the strongest classical methods could handle, according to Qunova’s CEO. He framed the run as the first time a quantum algorithm has met that same standard.&lt;/p&gt;
&lt;p&gt;In a preprint the company published with RIKEN researchers in September 2026, an earlier version of HI-VQE used 72 qubits for the same cluster. That run reached a ground-state energy of about -327.1757 Hartree against a DMRG reference of about -327.2396 Hartree, a gap of roughly 64 millihartree. It fell short of chemical accuracy, and needed about 66,249 node-hours on Fugaku to get even that close. Qunova says the November result needed only about 2,500 node-hours. That is a cut of roughly 300 times in computational cost versus its earlier quantum-diagonalisation approach, according to Qunova.&lt;/p&gt;
&lt;p&gt;The project involved IBM’s quantum unit and Japan’s RIKEN research centre, according to Qunova. Qunova says it was funded through NEDO, Japan’s state energy and industrial technology agency, under a programme backed by the country’s trade ministry. Qunova joined that NEDO-backed JHPC-quantum programme earlier in 2026 to deploy high-precision chemistry solvers for Japanese users.&lt;/p&gt;
</content:encoded><dc:creator>Mara Quinlan</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-FugakuSupercomputerSC19_wTN98.jpg" medium="image"/><category>computing</category></item><item><title>Which Quantum Advantage Claims Have Been Made, And Which Held Up?</title><link>https://explorequantum.com/article/which-quantum-advantage-claims-have-held-up/</link><guid isPermaLink="true">https://explorequantum.com/article/which-quantum-advantage-claims-have-held-up/</guid><description>Google, USTC and D-Wave have each said their quantum hardware beat the best classical computers. Classical teams answered most of those claims within years, sometimes months. Google&apos;s October 2025 Quantum Echoes result is the first built to be checked independently rather than only raced against.</description><pubDate>Thu, 12 Nov 2026 16:32:42 GMT</pubDate><content:encoded>&lt;p&gt;Several “quantum advantage” claims have been made since Google’s original Sycamore result in 2019. Classical computers have closed the gap on most of them, sometimes within months. The newest is different. Google’s “Quantum Echoes” result on its Willow chip, published in Nature on 22 October 2025, is the first the company designed so that other quantum computers can reproduce and check it. Earlier claims only raced a classical machine.&lt;/p&gt;
&lt;h2 id=&quot;the-first-claim-and-the-fastest-rebuttal&quot;&gt;The first claim, and the fastest rebuttal&lt;/h2&gt;
&lt;p&gt;Google published in Nature on 23 October 2019 that its 53-qubit Sycamore processor had sampled a random quantum circuit’s output in about 200 seconds. It said the fastest classical supercomputer of the time would need around 10,000 years for that task. That was the first claim of “quantum supremacy”: a case where a quantum computer does something no classical computer can realistically match.&lt;/p&gt;
&lt;p&gt;IBM’s rebuttal appeared a day before Google’s paper came out in Nature. According to IBM, researchers Edwin Pednault, John Gunnels, Dmitri Maslov and Jay Gambetta argued in an October 2019 blog post that an ideal classical simulation of the same circuit could run in about 2.5 days. That simulation would deliver far greater accuracy, using a supercomputer’s full RAM and hard-disk storage, they said. The original 10,000-year threshold, in their view, had not actually been crossed.&lt;/p&gt;
&lt;h2 id=&quot;chinas-larger-claim-and-the-algorithms-that-caught-up&quot;&gt;China’s larger claim, and the algorithms that caught up&lt;/h2&gt;
&lt;p&gt;USTC reported its own quantum-advantage claim in June 2021, later published in Physical Review Letters. Its 56-qubit Zuchongzhi processor sampled random circuits in about 1.2 hours. The team estimated the most powerful classical supercomputer would need at least 8 years for that task, 2 to 3 orders of magnitude harder than Sycamore’s result.&lt;/p&gt;
&lt;p&gt;Classical methods then went after both claims, with uneven results. In November 2021, physicists Feng Pan, Keyang Chen and Pan Zhang reported that a tensor-network algorithm run on a 512-GPU cluster solved the Sycamore sampling problem in about 15 hours. They estimated an exascale supercomputer could do the same task in a few dozen seconds, far below Google’s original 10,000-year estimate. A separate team went further in a June 2024 paper posted on arXiv. Using 1,432 Nvidia A100 GPUs, it generated 3 million uncorrelated samples of the Sycamore circuit in about 86 seconds, seven times faster than Google’s original experiment, while using about 13.7 kilowatt-hours of electricity. Between them, those two results closed most of the gap the 2019 Sycamore claim had opened. Zuchongzhi’s larger claim proved harder to match. In a paper posted on arXiv on 30 December 2021, Gleb Kalachev, Pavel Panteleev, PengFei Zhou and Man-Hong Yung described a bounded-fidelity classical sampling method. It produced 1 million low-fidelity samples of Zuchongzhi’s 56-qubit circuit on the Selene supercomputer, using 4,480 Nvidia A100 GPUs, in about 4 days. That narrowed the classical cost of USTC’s claim without closing it the way Sycamore’s claim was closed.&lt;/p&gt;
&lt;h2 id=&quot;a-logical-qubit-claim-undone-by-a-faster-algorithm&quot;&gt;A logical-qubit claim, undone by a faster algorithm&lt;/h2&gt;
&lt;p&gt;A team at Harvard, MIT and QuEra took a different route, with error correction rather than raw speed. According to Harvard and QuEra, it reported in Nature in December 2023 that its 48-logical-qubit neutral-atom processor ran instantaneous quantum polynomial-time circuits. Those circuits, the team said, would become rapidly harder to simulate classically as more circuit layers were added, part of a wider demonstration of error-corrected logical-qubit operations.&lt;/p&gt;
&lt;p&gt;That claim did not last long. According to IBM and IonQ, a classical algorithm built by IBM’s Dmitri Maslov and Sergey Bravyi with IonQ researchers computed an amplitude for the same circuit in about 0.0026 seconds in a February 2024 paper. That was roughly 1,000 times faster than the simulation time the Harvard/QuEra team had itself reported, challenging the idea that those circuits were classically intractable.&lt;/p&gt;
&lt;h2 id=&quot;an-annealers-claim-and-a-quick-classical-answer&quot;&gt;An annealer’s claim, and a quick classical answer&lt;/h2&gt;
&lt;p&gt;D-Wave took the next run at advantage with a different kind of hardware. According to D-Wave, it said in a paper published in Science on 12 March 2025 that its Advantage2 annealing prototype simulated the dynamics of programmable spin-glass magnetic materials in minutes. It said the Frontier supercomputer at Oak Ridge would need almost one million years for that task, and more electricity than the world uses in a year.&lt;/p&gt;
&lt;p&gt;Joseph Tindall and colleagues at the Flatiron Institute answered five days earlier than D-Wave’s own publication. According to the Flatiron Institute, they posted a paper on arXiv on 7 March 2025 describing a classical tensor-network method. It matched or beat D-Wave’s Advantage2 results on several of the same spin-glass lattices, verified on systems involving hundreds of qubits, disputing the quantum-supremacy claim before the Science paper was even out.&lt;/p&gt;
&lt;h2 id=&quot;what-makes-the-newest-claim-different&quot;&gt;What makes the newest claim different&lt;/h2&gt;
&lt;p&gt;Every earlier claim rested on the same bet: that no classical computer, however large, could keep up. In most cases, a better algorithm eventually closed most of that gap, or came close enough to make the claim debatable; Zuchongzhi’s claim only narrowed this way, without closing. Google’s Quantum Echoes result rests on a different kind of evidence. Running on its 105-qubit Willow chip, the algorithm measures out-of-time-order correlators about 13,000 times faster than the Frontier supercomputer could, according to Google. The company calls it the first “verifiable” quantum advantage for that reason: other quantum computers of similar quality can reproduce and cross-check the result themselves.&lt;/p&gt;
&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Claim&lt;/th&gt;
&lt;th&gt;Hardware&lt;/th&gt;
&lt;th&gt;Classical answer&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Sycamore, Oct 2019.&lt;/td&gt;
&lt;td&gt;53-qubit superconducting chip.&lt;/td&gt;
&lt;td&gt;Disputed within a day by IBM; largely closed by a 2024 GPU result.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Zuchongzhi, Jun 2021.&lt;/td&gt;
&lt;td&gt;56-qubit superconducting chip.&lt;/td&gt;
&lt;td&gt;Narrowed, not closed, by a December 2021 bounded-fidelity algorithm.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Harvard/QuEra IQP, Dec 2023.&lt;/td&gt;
&lt;td&gt;48 logical qubits, neutral atoms.&lt;/td&gt;
&lt;td&gt;Classical algorithm ran about 1,000 times faster than the team’s own reported simulation time.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;D-Wave Advantage2, Mar 2025.&lt;/td&gt;
&lt;td&gt;Quantum annealer.&lt;/td&gt;
&lt;td&gt;Matched by a classical tensor-network method days earlier.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Google Quantum Echoes, Oct 2025.&lt;/td&gt;
&lt;td&gt;105-qubit Willow chip.&lt;/td&gt;
&lt;td&gt;Designed to be cross-checked by other quantum hardware rather than raced against a classical one.&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p&gt;Reproducing the Quantum Echoes result, Google says, takes a quantum computer of comparable quality to Willow, not a classical challenger.&lt;/p&gt;
</content:encoded><dc:creator>Mara Quinlan</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-Measuring_a_qubit_leaves_no_room_for_error_Z1dBvg9.jpg" medium="image"/><category>computing</category></item><item><title>QuTech&apos;s Delft Spin-Qubit Team Wins Six-Year ERC Synergy Grant</title><link>https://explorequantum.com/article/qutech-four-researchers-win-erc-synergy-grant/</link><guid isPermaLink="true">https://explorequantum.com/article/qutech-four-researchers-win-erc-synergy-grant/</guid><description>Four QuTech researchers in Delft have won a six-year ERC Synergy Grant for project UniQ, aiming to build scalable germanium spin-qubit processors with built-in error correction, against the most competitive Synergy Grant call since it launched.</description><pubDate>Thu, 12 Nov 2026 14:13:06 GMT</pubDate><content:encoded>&lt;p&gt;Four Delft-based researchers at QuTech have won a six-year ERC Synergy Grant, according to QuTech. It funds their joint project, named UniQ, on scalable spin-qubit processors. Menno Veldhorst leads qubit architecture and operation, Fabio Sebastiano the on-chip electronics, Giordano Scappucci the semiconductor materials, and Lieven Vandersypen the system integration and testing. QuTech did not state the grant’s monetary value.&lt;/p&gt;
&lt;p&gt;The team’s plan is to combine germanium-based spin-qubit hardware with dedicated on-chip control electronics, according to QuTech. It also wants new ways of shuttling qubits within an array. Its stated end goal over the grant’s six-year term is to implement error correction on a scaled-up spin-qubit processor.&lt;/p&gt;
&lt;p&gt;Spin qubits are approaching a turning point, according to Menno Veldhorst, quoted by QuTech. Control over individual qubits keeps improving, Veldhorst said. The field is now also learning how to link and shuttle qubits between sites on a chip, he said.&lt;/p&gt;
&lt;p&gt;The award came out of the most competitive ERC Synergy Grant round since the call launched in 2012, according to the European Research Council. The ERC received 957 proposals for the 2026 call. Its 2026 Work Programme set aside EUR 500 million to fund an estimated 50 of them. That underlines how selective the Delft team’s win is.&lt;/p&gt;
&lt;p&gt;The four researchers now have six years to show whether germanium spin qubits can scale with error correction built in. That is the goal they have set for project UniQ.&lt;/p&gt;
</content:encoded><dc:creator>Owen Hartley</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-QuTech_Lab_Tour_(26819091690)_1v47co.jpg" medium="image"/><category>business</category></item><item><title>What Are Optical Atomic Clocks, And How Precise Are They?</title><link>https://explorequantum.com/article/what-are-optical-atomic-clocks-and-how-precise-are-they/</link><guid isPermaLink="true">https://explorequantum.com/article/what-are-optical-atomic-clocks-and-how-precise-are-they/</guid><description>Optical atomic clocks run on a near-petahertz transition, not caesium&apos;s microwave one, reaching top accuracy far faster. NIST&apos;s ytterbium clock has hit a stability record of 3.2 parts in 10^19 a day, and the tech spans chip-scale prototypes to a rack-mount clock trialled by the Royal Navy.</description><pubDate>Thu, 12 Nov 2026 10:19:19 GMT</pubDate><content:encoded>&lt;p&gt;According to NIST, an optical atomic clock ticks on an atomic transition at optical, near-petahertz frequency, rather than the microwave, gigahertz transition that caesium clocks use. NIST’s ytterbium version has reached a stability of about 3.2 parts in 10^19 when averaged over a day, a record for any atomic clock when NIST reported it in Nature on 28 November 2018. That ytterbium clock holds roughly a thousand atoms, trapped in pancake-shaped wells of laser light. A stability that fine means the clock would neither gain nor lose a second over tens of billions of years.&lt;/p&gt;
&lt;p&gt;Every atomic clock, optical or microwave, works the same basic way. It counts the fixed, natural oscillation of an atom, and uses that count to mark off seconds, instead of a pendulum or a quartz crystal. The difference between the two families is simply how fast that internal oscillation ticks. That difference turns out to matter a great deal for accuracy. According to NIST, caesium clocks need long averaging times to reach their best accuracy. Its NIST-F1 caesium fountain clock takes about five days, roughly 400,000 seconds. A ytterbium optical clock reaches comparable stability in about one second, NIST says. Settling on its best accuracy in a second rather than a working week lets researchers compare, tune and improve the clock far more quickly in the lab.&lt;/p&gt;
&lt;p&gt;That gain is why the body that defines the second wants to switch references. BIPM plans to redefine the SI second on an optical atomic transition, in place of the current caesium-microwave definition. Ytterbium-171 and strontium-87 are the leading candidates; both already feed into International Atomic Time as secondary representations of the second. BIPM anticipates adoption in 2030 or later, once clock stability, uncertainty and comparison techniques meet its roadmap criteria. The second has been redefined before, from an astronomical measure to the caesium transition. A move to an optical one would be the next step in the same long search for a steadier reference.&lt;/p&gt;
&lt;p&gt;Optical clocks are also shrinking, from laboratory instruments toward deployable hardware. NIST built a chip-scale optical atomic clock using integrated photonics and metasurface optics. It is roughly one-thousandth the size of NIST’s record-setting laboratory ytterbium clock at JILA, only a few inches on a side excluding its lasers. NIST lists potential uses including timing for power grids, data centres and financial markets, gravity-sensing quantum sensors, geodesy and navigation, and space-based physics experiments. Power grids and financial exchanges already depend on precise, synchronised timing to order transactions and keep electricity flowing smoothly. That is why a clock small enough to leave the laboratory matters as much as one that is merely more stable.&lt;/p&gt;
&lt;p&gt;A commercial version is already further along the deployment path. Infleqtion’s Tiqker optical atomic clock fits in a roughly 30-litre, 30-kilogram 3U rack-mount unit, about a tenth the size of a hydrogen maser. The company says it delivers hydrogen-maser-like short-term stability, outperforming caesium-beam references, rubidium standards and chip-scale atomic clocks on short-to-mid timescales. In October 2025, Infleqtion says Tiqker was trialled aboard the Royal Navy’s Excalibur autonomous underwater vehicle, to advance GPS-free navigation. The trial let the platform hold accurate timing and position for longer when it has no satellite signal. Lose the satellite link, and an autonomous vehicle normally loses track of time and position together. A clock stable enough to hold both on its own matters to any platform that operates underwater, underground or out of satellite reach.&lt;/p&gt;
</content:encoded><dc:creator>Mara Quinlan</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-Experimental_Atomic_Clock__ytterbium_atoms_(5941021758)_1CM4t3.jpg" medium="image"/><category>sensing</category></item><item><title>Xanadu&apos;s Q3 2026 Revenue Rises 32% as Net Loss Widens to $32.2 Million</title><link>https://explorequantum.com/article/xanadu-q3-2026-results-cash-government-funding/</link><guid isPermaLink="true">https://explorequantum.com/article/xanadu-q3-2026-results-cash-government-funding/</guid><description>Xanadu&apos;s Q3 2026 revenue rose 32% to about $1.3 million while its net loss widened to $32.2 million on higher R&amp;D spending. The photonic quantum computer maker held $286.5 million in cash and secured about $140 million in Canadian government financing for a manufacturing facility.</description><pubDate>Thu, 12 Nov 2026 10:18:00 GMT</pubDate><content:encoded>&lt;p&gt;Toronto-based photonic quantum computing company Xanadu reported third-quarter 2026 revenue of about $1.3 million, according to Xanadu’s Q3 2026 results. That is up roughly 32% from the same quarter a year earlier. It held a cash balance of about $286.5 million as of 30 September 2026.&lt;/p&gt;
&lt;p&gt;That growth came alongside a wider net loss, according to Xanadu’s Q3 2026 results. The loss rose to about $32.2 million, from about $20.3 million a year earlier. Xanadu says the increase reflects higher R&amp;amp;D spending, which climbed to roughly $25.6 million from about $15.3 million. General and administrative expenses rose to about $9.2 million, and capital expenditure was about $8.4 million.&lt;/p&gt;
&lt;p&gt;That spending is funded partly through non-dilutive sources, Xanadu’s Q3 2026 results show. The company has secured roughly $140 million in non-dilutive financing through a loan from a Canadian federal fund, earmarked for a manufacturing facility. It also has provisional approval for up to $10 million more under an extension of the DARPA Quantum Benchmarking Initiative. About $232.8 million remains available to it under an equity facility.&lt;/p&gt;
&lt;p&gt;The quarter also brought a roster of partnerships, Xanadu’s Q3 2026 results show. These include work with ASML on reducing waveguide loss, with Bluefors on compact cooling systems, and with GlobalFoundries on scaling a silicon nitride photonics platform. The company also launched a classical-computing control framework with AMD. ASML’s involvement adds a European name to a list otherwise built around US and Canadian ties. Waveguide loss, where light signals weaken as they cross chip-scale photonic circuits, is one of the central engineering hurdles the photonic approach to quantum computing has to clear. Xanadu also lists workforce-development ties from the same quarter to the Royal Bank of Canada, Booz Allen and Lockheed Martin.&lt;/p&gt;
</content:encoded><dc:creator>Owen Hartley</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-Toronto_-_ON_-_Skyline_bei_Nacht_Z1OE8h8.jpg" medium="image"/><category>business</category></item><item><title>What Are Quantum Sensors, and Where Are They Already Used?</title><link>https://explorequantum.com/article/what-are-quantum-sensors-and-where-are-they-used/</link><guid isPermaLink="true">https://explorequantum.com/article/what-are-quantum-sensors-and-where-are-they-used/</guid><description>A quantum sensor uses atoms, electron spins or superconducting circuits to measure what classical instruments cannot, NIST says. Quantum magnetometers already scan brains and hearts, and portable atomic clocks already find offshore oil.</description><pubDate>Thu, 12 Nov 2026 05:22:05 GMT</pubDate><content:encoded>&lt;p&gt;A quantum sensor uses a small quantum system as the part that touches the world, according to NIST. That system is most often atoms, electron spins or superconducting circuits. It measures a physical quantity in a way classical physics alone cannot match. Most quantum sensors follow the same three-step loop, NIST says: prepare a known quantum state, let the world nudge it, then read out the change. NIST lists three advantages this gives over classical instruments. Sensitivity is one, down to detecting the energy of a single photon. Self-calibration is another, because atoms of a given type are always identical, unlike mechanical parts. Small size follows, since the devices are built around atoms or microscopic circuits.&lt;/p&gt;
&lt;h2 id=&quot;a-technology-older-than-the-word-quantum-sensor&quot;&gt;A technology older than the word “quantum sensor”&lt;/h2&gt;
&lt;p&gt;NIST traces the practical history of quantum sensing through four milestones. The atomic clock came first, invented in 1949. Magnetic field sensors using atomic spins followed from the 1950s. Superconducting quantum interference devices, or SQUIDs, came from the 1960s. Magnetic resonance imaging, which uses quantum spin to image the body, dates to the 1970s. The label “quantum sensor” is newer than the hardware it describes.&lt;/p&gt;
&lt;h2 id=&quot;reading-magnetic-fields-three-ways&quot;&gt;Reading magnetic fields three ways&lt;/h2&gt;
&lt;p&gt;Quantum magnetometers are far more sensitive than classical devices such as compasses or flux-gate magnetometers, NIST says. They come in three main kinds. SQUIDs use a superconducting loop whose current oscillates as a magnetic field is applied. Atomic magnetometers trap billions of caesium or rubidium atoms and align their electron spins with polarised laser light. They then measure how a magnetic field deflects those spins. Nitrogen-vacancy, or NV, centre magnetometers read magnetic-field effects on electron spins in diamond defects with laser light.&lt;/p&gt;
&lt;p&gt;These devices are already in clinical and field use. Quantum magnetometers power magnetoencephalography, which images the brain’s magnetic activity, NIST says. It helps diagnose epilepsy, multiple sclerosis, Alzheimer’s disease and schizophrenia. NIST itself first demonstrated the equivalent technique for the heart, magnetocardiography, in 2014. That demonstration was for monitoring a fetal heartbeat. Beyond medicine, NIST points to quantum magnetometers already helping locate underground minerals and buried archaeological sites. They also support dark-matter research and microwave astronomy observations. An emerging use is measuring Earth’s magnetic field for GPS-free navigation.&lt;/p&gt;
&lt;h2 id=&quot;turning-falling-atoms-into-a-gravity-sensor&quot;&gt;Turning falling atoms into a gravity sensor&lt;/h2&gt;
&lt;p&gt;Quantum gravimeters work differently. According to NIST, they cool atoms to near absolute zero to bring out their wave nature. A laser pulse then splits each atom’s wave into two paths. Those paths fall along slightly different trajectories, where gravity differs by a tiny amount. A second laser pulse recombines the waves into an interference pattern that encodes that gravitational difference, NIST says.&lt;/p&gt;
&lt;p&gt;One use is already commercial. Companies use portable atomic clocks positioned above the seabed to detect oil deposits, NIST says. A deposit’s slightly stronger gravitational pull makes a clock above it tick marginally more slowly than a reference clock elsewhere. That effect is predicted by Einstein’s general relativity. NIST lists further, still-emerging gravimeter uses. They include tracking lava movement inside volcanoes and finding mineral or oil deposits. Others are monitoring how much water an aquifer holds and how fast it recharges, and detecting hidden tunnels or voids under infrastructure. NIST also names sharpening the definition of the kilogram through more accurate local gravity readings, and GPS-independent navigation for submarines and deep-space vehicles. One quantum gravimeter unit has already been launched into space, NIST notes.&lt;/p&gt;
&lt;h2 id=&quot;clocks-as-physics-tools&quot;&gt;Clocks as physics tools&lt;/h2&gt;
&lt;p&gt;According to NIST, an atomic clock’s tick rate depends on quantum energy levels. General relativity says those levels run slower in stronger gravity. So scientists have used atomic clocks as gravity sensors for precision tests of Einstein’s theory, NIST says. That has included detecting tiny differences in ticking rates between atoms separated by just a millimetre in height. More recently it has meant placing ultra-accurate clocks on mountains and aboard the International Space Station.&lt;/p&gt;
&lt;h2 id=&quot;europes-research-pipeline&quot;&gt;Europe’s research pipeline&lt;/h2&gt;
&lt;p&gt;Germany’s national aerospace research centre, DLR, runs a Quantum Sensing department. It develops inertial sensors based on atom interferometry, the same wave-splitting principle behind quantum gravimeters. The aim is high-precision measurement of rotation and acceleration with long-term stability that avoids the drift of conventional mechanical sensors. DLR names terrestrial and satellite gravimetry and gradiometry, Earth observation, and navigation and exploration as the application areas it is developing the technology for.&lt;/p&gt;
&lt;h2 id=&quot;what-to-watch&quot;&gt;What to watch&lt;/h2&gt;
&lt;p&gt;Medical imaging, oil and mineral survey, and relativity tests already rest on quantum sensors in service today. The gravimeter applications NIST and DLR describe are the next ones to move from research programme to routine use. They range from volcano monitoring to GPS-free navigation for submarines and spacecraft.&lt;/p&gt;
</content:encoded><dc:creator>Mara Quinlan</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-Optical_atomic_clock_(5881387757)_Z1Uzswp.jpg" medium="image"/><category>sensing</category></item><item><title>Rigetti&apos;s 8-K Confirms Commerce Department Took Equity Stake Tied to $100M CHIPS Act Deal</title><link>https://explorequantum.com/article/rigetti-chips-act-commerce-equity-stake/</link><guid isPermaLink="true">https://explorequantum.com/article/rigetti-chips-act-commerce-equity-stake/</guid><description>Rigetti Computing&apos;s Q3 2026 SEC filing discloses that the U.S. Department of Commerce received a minority equity stake as a condition of the $100 million CHIPS Act funding agreement the two sides signed in September.</description><pubDate>Thu, 12 Nov 2026 02:11:34 GMT</pubDate><content:encoded>&lt;p&gt;The U.S. Department of Commerce took a minority, non-controlling equity stake in Rigetti Computing, according to Rigetti. It came as a condition of a federal funding agreement. Rigetti disclosed the stake in its third-quarter 2026 earnings release. The filing did not state what percentage of the company the government now holds.&lt;/p&gt;
&lt;p&gt;The stake traces back to a deal Rigetti signed with Commerce in September 2026, according to Rigetti. That deal covers up to $100 million in CHIPS Act funding. The money is earmarked for three research programmes: miniaturised readout electronics, expanded cryogenic capacity, and fabrication of high-connectivity chip architectures.&lt;/p&gt;
&lt;p&gt;Rigetti filed a Form 8-K with the SEC on 10 November 2026, under Items 2.02 and 9.01. It furnished the earnings release as an exhibit, rather than filing it as part of the form itself.&lt;/p&gt;
&lt;p&gt;Rigetti has not said when the stake will be finalised, what percentage of the company it covers, or what rights it carries beyond being non-controlling.&lt;/p&gt;
</content:encoded><dc:creator>Owen Hartley</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-United_States_Department_of_Commerce_-_Herbert_C._Hoover_Building_-_Washington_D.C._1c4NYG.jpg" medium="image"/><category>business</category></item><item><title>Rigetti Hits 99.5% Gate Fidelity as D-Wave, Horizon Quantum and QCi Post Q3 2026 Results</title><link>https://explorequantum.com/article/q3-2026-quantum-earnings-dwave-rigetti-horizon-qci/</link><guid isPermaLink="true">https://explorequantum.com/article/q3-2026-quantum-earnings-dwave-rigetti-horizon-qci/</guid><description>Rigetti Computing reported a 99.5% two-qubit gate fidelity milestone alongside its Q3 2026 results, as D-Wave posted 21% revenue growth and a finalized $100 million CHIPS Act deal, Horizon Quantum disclosed a $27 million Entropica Labs acquisition, and QCi&apos;s revenue jumped on recent acquisitions.</description><pubDate>Wed, 11 Nov 2026 18:59:16 GMT</pubDate><content:encoded>&lt;p&gt;Rigetti Computing reached a median two-qubit gate fidelity of 99.5% on its Cepheus-1-108Q processor, according to Rigetti. Chief executive Subodh Kulkarni tied the milestone to Rigetti’s roadmap toward a system of more than 300 qubits by 2028. Further out, Rigetti is targeting a machine of more than 1,000 qubits, with even higher fidelity and faster gates. The milestone came alongside Rigetti’s third-quarter 2026 results. The full-stack quantum-classical computing firm reported quarterly revenue of about $1.4 million. Nine-month revenue reached roughly $10.9 million, up about 110% from the first nine months of 2025. Rigetti said it still expects full-year revenue near $20 million. Rigetti said the company held about $497.1 million in cash, cash equivalents and investments, with no debt, at quarter’s end. Its quarterly operating loss came in at roughly $33.2 million.&lt;/p&gt;
&lt;p&gt;D-Wave Quantum, the Canadian-American annealing and gate-model quantum computer maker, reported third-quarter revenue of about $4.5 million. That is up roughly 21% from the same quarter a year earlier, which D-Wave attributed to growth in professional services and its quantum-computing-as-a-service business. D-Wave said its remaining performance obligations, a measure of contracted future revenue, jumped to about $38.3 million at the end of the quarter. That is up roughly 1,221% from a year earlier, while nine-month bookings for 2026 climbed to about $37.4 million. The company also finalized an agreement with the United States Department of Commerce under the CHIPS and Science Act, D-Wave said. The deal gives D-Wave access to up to $100 million in funding, paid out in installments tied to milestones. D-Wave posted a third-quarter net loss of about $48.2 million, the company reported. Both outlets covering the results described that as a large improvement on the roughly $140 million loss D-Wave reported for the same quarter last year, alongside a rise in operating expenses tied to growth spending. D-Wave reported holding roughly $503 million in cash and investments at the close of the quarter. It has pointed to that cushion as funding for continued expansion of both its annealing and gate-model hardware lines.&lt;/p&gt;
&lt;p&gt;Horizon Quantum Holdings, a quantum software infrastructure company, reported third-quarter net income of about $33.4 million, according to Horizon Quantum. The company said that was driven mainly by a $43.1 million non-cash gain on warrant valuations, rather than operations. Its underlying operating loss was about $10.5 million, and cash fell to roughly $102.8 million. Alongside its results, Horizon Quantum disclosed a non-binding letter of intent to acquire Singapore-based quantum software company Entropica Labs for about $27 million in shares.&lt;/p&gt;
&lt;p&gt;According to Quantum Computing Inc., the company’s third-quarter revenue rose to about $9.5 million, from roughly $384,000 a year earlier. Growth came primarily from three businesses it recently acquired: Luminar Semiconductor, NuCrypt and NHanced Semiconductors. The company did not break out the three acquisitions’ individual revenue contributions in the release.&lt;/p&gt;
&lt;p&gt;Taken together, the four reports show a gap between quantum-computing revenue and the cash these companies hold. D-Wave’s $4.5 million and Rigetti’s $1.4 million in quarterly revenue each sit well below the hundreds of millions of dollars the two companies separately reported holding in cash and investments. Horizon Quantum’s reported net income, by contrast, came almost entirely from a non-cash accounting gain, rather than from sales.&lt;/p&gt;
</content:encoded><dc:creator>Owen Hartley</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-Superconducting_Metamaterial_Chip_002_2sfH0E.jpg" medium="image"/><category>business</category></item><item><title>Quantinuum and RIKEN Run Two Quantum Processors Against Fugaku at Once</title><link>https://explorequantum.com/article/quantinuum-riken-fugaku-multi-qpu/</link><guid isPermaLink="true">https://explorequantum.com/article/quantinuum-riken-fugaku-multi-qpu/</guid><description>Quantinuum, the University of Tsukuba and RIKEN linked Japan&apos;s Fugaku supercomputer to two quantum processors at the same time, an on-site superconducting chip and a trapped-ion system 500 km away, and used the combination to cut a benchmark&apos;s energy error from 2.47 to 0.61.</description><pubDate>Wed, 11 Nov 2026 15:02:25 GMT</pubDate><content:encoded>&lt;p&gt;Quantinuum, the University of Tsukuba and RIKEN’s Center for Computational Science have run a workflow that ties Japan’s Fugaku supercomputer to two different quantum processors at once, according to a Quantinuum-authored paper. One chip, an on-site IBM superconducting processor, sat next to Fugaku in Kobe. The other, Quantinuum’s trapped-ion Reimei system, sat 500 km away in Wako. Quantinuum’s Tierkreis middleware, software that routes jobs between classical and quantum hardware, coordinated the split.&lt;/p&gt;
&lt;p&gt;According to the same Quantinuum-authored paper, the team tested the setup on a 16-site two-dimensional transverse-field Ising model, a standard benchmark for simulating magnetic materials. The nearby superconducting chip ran 500 variational iterations, taking 500 shots per measurement, to find good parameters. It then handed those parameters to the remote trapped-ion system for a slower but more accurate phase-estimation step that took 300 shots. That handoff cut the benchmark’s energy error from 2.47 to 0.61.&lt;/p&gt;
&lt;p&gt;The trapped-ion side of that split has since been upgraded. RIKEN moved Reimei from Quantinuum’s 20-qubit System Model H1 to the 56-qubit System Model H2, a trapped-ion processor with all-to-all qubit connectivity, according to Quantinuum. The new system keeps its direct, low-latency link into Fugaku’s compute partitions. It is aimed at larger computational-chemistry problems such as drug discovery and materials science.&lt;/p&gt;
&lt;p&gt;SoftBank has joined Quantinuum and RIKEN in expanding their partnership to commercialise hybrid quantum-AI supercomputing in Japan, according to Quantinuum. The three partners want to build a domestically built, fault-tolerant quantum computer. The plan combines the upgraded Reimei H2 with RIKEN’s ROQUO GPU cluster and Fugaku. SoftBank is assessing a direct link between quantum hardware and its own commercial AI data centres. Announcing the expansion, Quantinuum chief executive Rajeeb Hazra called RIKEN’s continued use of Quantinuum hardware a validation of its technology roadmap. He said the Reimei upgrade widens what the two groups can explore together across electronics, energy and life sciences. RIKEN’s Mitsuhisa Sato said the upgraded system lets researchers take on more complex and larger-scale scientific problems.&lt;/p&gt;
&lt;p&gt;A Quantinuum-authored paper separately reports an end-to-end hybrid workflow that Quantinuum and RIKEN ran on Fugaku and Reimei. The workflow calculated biomolecular excited-state energies, orchestrating the supercomputer and the quantum coprocessor directly through an API rather than through Tierkreis. RIKEN and Quantinuum say the next step is the fault-tolerant quantum computer the expanded partnership is working toward.&lt;/p&gt;
</content:encoded><dc:creator>Mara Quinlan</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-RIKEN_R-CCS_Fugaku_1Q1DUe.jpg" medium="image"/><category>computing</category></item><item><title>What Is Quantum Key Distribution, and Where Is It Already Used?</title><link>https://explorequantum.com/article/what-is-quantum-key-distribution/</link><guid isPermaLink="true">https://explorequantum.com/article/what-is-quantum-key-distribution/</guid><description>Quantum key distribution encodes an encryption key in single photons so eavesdropping leaves a detectable trace, NIST says. All 27 EU states now take part in a terrestrial and satellite QKD network, and China and South Korea already run links in production.</description><pubDate>Wed, 11 Nov 2026 10:16:21 GMT</pubDate><content:encoded>&lt;p&gt;According to NIST, the US National Institute of Standards and Technology, quantum key distribution lets two parties agree on a secret encryption key. They send it encoded in individual photons rather than over an ordinary data channel. Each photon’s polarisation carries one random bit. The sender picks a random mode and value for every photon, and the receiver measures each one in an independently chosen mode. Only the photons measured in the mode the sender actually used become part of the shared key.&lt;/p&gt;
&lt;h2 id=&quot;why-it-is-considered-secure&quot;&gt;Why it is considered secure&lt;/h2&gt;
&lt;p&gt;QKD’s security rests on a basic quantum fact: measuring a particle disturbs it, NIST says. An eavesdropper who intercepts the photons introduces detectable errors. The two parties compare a sample of their measurements, and discard the key if the error rate is too high. NIST’s own prototype system sends 312 million digital values per second over the quantum channel.&lt;/p&gt;
&lt;p&gt;The idea goes back to 1984, when Charles Bennett and Gilles Brassard proposed the BB84 protocol, encoding key bits in the polarisation states of single photons, according to Wikipedia. BB84 remains the best-known QKD scheme and the basis most later fibre and satellite systems build on.&lt;/p&gt;
&lt;h2 id=&quot;where-it-already-runs&quot;&gt;Where it already runs&lt;/h2&gt;
&lt;p&gt;All 27 EU member states now take part in the European Quantum Communication Infrastructure, the European Commission says. It combines terrestrial fibre links between government, data-centre and critical-infrastructure sites with a satellite segment. A first terrestrial implementation phase ran in 2023-2025. The Commission’s Eagle-1 prototype QKD satellite is due to launch in late 2027.&lt;/p&gt;
&lt;p&gt;China runs the largest deployment reported so far. According to USTC, the University of Science and Technology of China, its integrated network has a 2,000-kilometre fibre backbone linking Beijing, Jinan, Hefei and Shanghai. That fibre line joins a 2,600-kilometre satellite link between ground stations in Xinglong and Nanshan via the Micius satellite. The combined reach is over 4,600 kilometres, reported in Nature in January 2021. USTC says the network serves more than 150 industrial users, including banks, power grids and government agencies.&lt;/p&gt;
&lt;p&gt;South Korean carrier SK Telecom says it applied QKD to the Seoul-Daejeon section of its commercial 5G and LTE network in 2019, using technology from Geneva-based ID Quantique. That section is the most heavily trafficked stretch of its backbone in the country. SK Telecom has since led several international standardisation proposals on QKD at the ITU-T.&lt;/p&gt;
&lt;h2 id=&quot;what-governments-say-about-its-limits&quot;&gt;What governments say about its limits&lt;/h2&gt;
&lt;p&gt;Deployment is not the same as official endorsement. According to the NCSC, the UK’s National Cyber Security Centre, it does not endorse QKD for any government or military application. It advises against relying on QKD alone for business-critical networks, including critical national infrastructure. NCSC points to QKD’s lack of built-in authentication, the specialised hardware it needs, and the trusted relays long-distance links require. It recommends quantum-safe cryptographic algorithms instead as the primary defence against quantum computers.&lt;/p&gt;
&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Deployment&lt;/th&gt;
&lt;th&gt;Scale&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;EuroQCI (EU, 27 member states)&lt;/td&gt;
&lt;td&gt;Terrestrial fibre phase run 2023-2025; Eagle-1 satellite due late 2027.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;China (USTC)&lt;/td&gt;
&lt;td&gt;2,000 km fibre backbone plus 2,600 km satellite link, 4,600 km combined, serving 150+ industrial users.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;South Korea (SK Telecom)&lt;/td&gt;
&lt;td&gt;Seoul-Daejeon section of its commercial 5G/LTE backbone, since 2019.&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p&gt;Live deployment has outpaced that guidance. NCSC’s position against relying on QKD alone for critical infrastructure stands, even as EuroQCI’s satellite segment moves toward its 2027 launch.&lt;/p&gt;
</content:encoded><dc:creator>Mara Quinlan</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-High-Speed_Quantum_Key_System_Wins_Research_Award_(5884514398)_1fdupu.jpg" medium="image"/><category>networking</category></item><item><title>QuiX Quantum Leads Dutch Mission to Build a 100-Qubit Photonic Quantum Computer</title><link>https://explorequantum.com/article/quix-quantum-photonic-100-qubit/</link><guid isPermaLink="true">https://explorequantum.com/article/quix-quantum-photonic-100-qubit/</guid><description>QuiX Quantum and seven Dutch partners have started a three-year, PhotonDelta-funded mission to build the manufacturing and control technology for a universal photonic quantum computer with more than 100 qubits.</description><pubDate>Wed, 11 Nov 2026 10:11:02 GMT</pubDate><content:encoded>&lt;p&gt;QuiX Quantum is leading a Dutch national consortium building the manufacturing and architectural blocks for a photonic quantum computer with more than 100 qubits, the company says. The Photonic Quantum Computing Mission runs from January 2026 through December 2028. It is funded under Phase II of the PhotonDelta National Growth Fund programme. A photonic quantum computer encodes qubits in particles of light, not in trapped ions or superconducting circuits. That approach can run at room temperature, but it needs very low loss across its optical parts to scale up.&lt;/p&gt;
&lt;p&gt;The consortium groups QuiX Quantum with imec, PHIX, MicroAlign, the University of Twente, Eindhoven University of Technology, Leiden University and Bruco-IC. Together they cover every step from photon generation and waveguide fabrication to fiber-to-chip packaging, detection and feed-forward control electronics. QuiX Quantum, a Dutch-German maker of photonic quantum processors, leads the group.&lt;/p&gt;
&lt;p&gt;Two targets set the mission’s technical bar, according to QuiX Quantum. The first cuts control-electronics latency, the delay between measuring a photon and acting on that measurement. It should fall below 10 nanoseconds, down from about 150 nanoseconds today. The second target cuts fiber-to-chip coupling loss, the light lost where an optical fibre meets the chip, to around 0.1 decibels per facet. The consortium says both gains are needed to scale photonic processors past 100 qubits.&lt;/p&gt;
&lt;p&gt;QuiX Quantum CEO Stefan Hengesbach called the grant a strong vote of confidence in the Dutch photonics ecosystem and in the company’s roadmap. He said scalable, fault-tolerant quantum computing cannot be built by one company alone. He framed the consortium as the way the Netherlands pools photonics manufacturing and quantum expertise toward that goal.&lt;/p&gt;
&lt;p&gt;QuiX Quantum sets out twelve deliverables across the photonic stack for the three-year mission. imec takes the ultra-low-loss silicon nitride waveguides, Twente the single-photon sources, and Eindhoven the on-chip detectors and inverse-designed components. Leiden builds the photon-number-resolving detectors, and Bruco-IC the sub-10-nanosecond control chips meant to hit that latency target. The partners aim for loss-tolerant logical-qubit demonstrations, where several physical qubits are combined to suppress errors, between 2028 and 2030. They place commercially deployable fault-tolerant systems beyond 2030.&lt;/p&gt;
&lt;p&gt;The next marker to watch is whether the consortium reports progress on the latency and coupling-loss targets before the mission ends.&lt;/p&gt;
</content:encoded><dc:creator>Mara Quinlan</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-Universiteit_Twente_Hogekamp_closeup_pCp8f.jpg" medium="image"/><category>computing</category></item><item><title>How Does a Quantum Network Work, and How Far Along Is the Quantum Internet?</title><link>https://explorequantum.com/article/how-does-a-quantum-network-work/</link><guid isPermaLink="true">https://explorequantum.com/article/how-does-a-quantum-network-work/</guid><description>A quantum network moves qubits instead of bits through repeater nodes, says QuTech. Test links so far reach about 124 miles of fibre in the Chicago area and 17.6 km in New York, far short of a working quantum internet.</description><pubDate>Tue, 10 Nov 2026 16:28:11 GMT</pubDate><content:encoded>&lt;p&gt;A quantum network carries information encoded in qubits, not ordinary bits, according to QuTech and the University of Chicago. That lets it link quantum computers and sensors together, pool them into a larger computing cluster, and enable communication whose privacy is guaranteed by the laws of physics. The resulting “quantum internet” will not replace today’s internet, but add those capabilities on top of it, the University of Chicago says. Today that network is still a handful of short test links, not anything spanning a continent.&lt;/p&gt;
&lt;h2 id=&quot;how-the-signal-gets-from-a-to-b&quot;&gt;How the signal gets from A to B&lt;/h2&gt;
&lt;p&gt;Quantum repeaters, which relay entanglement over longer distances, are a basic component a quantum network needs to scale beyond two directly linked machines, QuTech says. They have to pass along entangled particles without collapsing the fragile quantum state that carries the information, which is far harder than copying and forwarding an ordinary signal. In 2021, QuTech reported it had built the first multi-node quantum network, connecting three quantum processors.&lt;/p&gt;
&lt;h2 id=&quot;europes-effort-the-quantum-internet-alliance&quot;&gt;Europe’s effort: the Quantum Internet Alliance&lt;/h2&gt;
&lt;p&gt;QuTech, a founding partner of the Quantum Internet Alliance (QIA), coordinates the group of about forty of Europe’s leading academic, industrial and research organisations, according to QuTech and TNO. It is running a seven-year programme to build what QuTech calls the world’s first full-stack prototype quantum internet network. TNO, QuTech’s partner in the alliance, says the programme’s overall goal is to link two metropolitan-scale quantum networks over a long-distance fibre backbone using quantum repeaters. Its first phase, running 3.5 years from October 2022 on a budget of 24 million euros, aims only to identify pilot use cases for early-stage quantum internet technology, TNO says.&lt;/p&gt;
&lt;h2 id=&quot;the-united-states-a-national-blueprint-and-a-124-mile-testbed&quot;&gt;The United States: a national blueprint and a 124-mile testbed&lt;/h2&gt;
&lt;p&gt;The US Department of Energy unveiled a national blueprint strategy for a quantum internet in July 2020, according to the department. It said its 17 national laboratories would serve as the backbone of the future network. The blueprint lays out four progressive stages, from verifying secure quantum protocols over existing fibre to eventually linking states using quantum repeaters. At the time, the department said a prototype was within reach over the following decade.&lt;/p&gt;
&lt;p&gt;Part of that effort now runs in Chicago. The University of Chicago says its Chicago-area quantum network, built with Argonne National Laboratory, reached 124 miles of optical fibre across six nodes after a June 2022 extension. That is an early step toward an entanglement-based quantum internet, the university says. Its researchers estimate interstate quantum networks could be established within the United States in the next 10 to 15 years.&lt;/p&gt;
&lt;h2 id=&quot;a-commercial-demonstration-in-new-york-city&quot;&gt;A commercial demonstration in New York City&lt;/h2&gt;
&lt;p&gt;Qunnect announced in February 2026 that it and Cisco had demonstrated entanglement swapping over 17.6 km of deployed commercial telecom fibre between Brooklyn and Manhattan, outlets report. Entanglement swapping is the technique that lets two particles become linked even though they never interacted directly. The system reached swap rates of about 5,400 entangled photon pairs per hour over that fibre, with better than 99% polarisation fidelity. Qunnect says it was the first such demonstration on deployed metro-scale fibre using a commercial quantum-networking system, a step toward linking quantum computers and data centres over existing telecom infrastructure.&lt;/p&gt;
&lt;h2 id=&quot;where-the-test-networks-stand&quot;&gt;Where the test networks stand&lt;/h2&gt;
&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Effort&lt;/th&gt;
&lt;th&gt;Who runs it&lt;/th&gt;
&lt;th&gt;Scale so far&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Quantum Internet Alliance&lt;/td&gt;
&lt;td&gt;QuTech, TNO and about 40 partners in Europe.&lt;/td&gt;
&lt;td&gt;First phase, from October 2022, budget EUR 24 million, aims to identify pilot use cases. Seven-year programme overall aims to link two metro-scale networks over a long-distance backbone.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;National quantum internet blueprint&lt;/td&gt;
&lt;td&gt;United States Department of Energy, 17 national labs.&lt;/td&gt;
&lt;td&gt;Four-stage roadmap set in 2020. A prototype was then seen as feasible within a decade.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Chicago-area testbed&lt;/td&gt;
&lt;td&gt;University of Chicago and Argonne National Laboratory.&lt;/td&gt;
&lt;td&gt;124 miles of fibre across six nodes, after a June 2022 extension. Interstate links estimated at 10-15 years out.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Metro-fibre entanglement swap&lt;/td&gt;
&lt;td&gt;Qunnect and Cisco.&lt;/td&gt;
&lt;td&gt;17.6 km of deployed commercial fibre in New York City, announced February 2026.&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p&gt;None of these efforts yet links more than a handful of nodes over metropolitan distances. Next to watch is whether the Quantum Internet Alliance reaches its programme goal of linking two metro-scale networks over a long-distance backbone; the Chicago testbed’s interstate timeline is the other date to track.&lt;/p&gt;
</content:encoded><dc:creator>Mara Quinlan</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-QuTech_Lab_Tour_(26819085560)_1VGEyp.jpg" medium="image"/><category>networking</category></item><item><title>Horizon Quantum Links Triple Alpha IDE to OQC&apos;s 32-Qubit Toshiko Processor for Pulse-Level Control</title><link>https://explorequantum.com/article/horizon-quantum-oqc-pulse-level-integration/</link><guid isPermaLink="true">https://explorequantum.com/article/horizon-quantum-oqc-pulse-level-integration/</guid><description>Horizon Quantum and Oxford Quantum Circuits have connected Horizon&apos;s Triple Alpha development environment directly to the control plane of OQC&apos;s 32-qubit Toshiko processor, giving users pulse-level access to qubit parameters such as T1 and T2 from inside the IDE.</description><pubDate>Tue, 10 Nov 2026 06:14:36 GMT</pubDate><content:encoded>&lt;p&gt;Horizon Quantum Computing and Oxford Quantum Circuits (OQC) have completed a pulse-level integration between Horizon’s Triple Alpha development environment and OQC’s Toshiko processor, according to Horizon Quantum and OQC. The link runs through a web API wired directly into Toshiko’s control plane. It lets users run pulse scheduling, qubit characterisation and automated gate calibration from inside Triple Alpha, instead of switching between separate toolchains.&lt;/p&gt;
&lt;p&gt;Toshiko is OQC’s 32-qubit superconducting processor, built on the company’s own 3D coaxial Coaxmon qubit design. Triple Alpha is Horizon’s IDE for building quantum programs.&lt;/p&gt;
&lt;p&gt;With the new link, researchers can read Toshiko’s physical qubit parameters directly from Triple Alpha, according to Horizon Quantum and OQC. That includes each qubit’s relaxation time (T1) and phase coherence (T2). Users can also adjust the pulse envelopes that shape single- and two-qubit operations, and deploy custom dynamical decoupling sequences. That tunes native gate fidelities from inside the IDE, rather than a separate control-system toolchain.&lt;/p&gt;
&lt;p&gt;Horizon Quantum chief executive Joe Fitzsimons said the integration lets users interact with the physics of OQC’s processors, rather than only the data they produce, according to Horizon Quantum and OQC. OQC chief product officer Simon Phillips said bringing OQC’s pulse-level capabilities into Triple Alpha gives developers more control. That control covers characterising qubits, calibrating gates and building quantum programs within a single environment.&lt;/p&gt;
</content:encoded><dc:creator>Mara Quinlan</dc:creator><media:content url="https://explorequantum.com/_astro/pexels-photo-30547566_6OOgw.jpg" medium="image"/><category>computing</category></item><item><title>IonQ Signs Carahsoft as Public Sector Distributor for Its Quantum Technology</title><link>https://explorequantum.com/article/ionq-carahsoft-public-sector-distribution/</link><guid isPermaLink="true">https://explorequantum.com/article/ionq-carahsoft-public-sector-distribution/</guid><description>Trapped-ion computing firm IonQ has made government IT distributor Carahsoft its public sector channel for quantum computing, networking, sensing and security products, reaching federal, state, local and higher-education buyers.</description><pubDate>Mon, 09 Nov 2026 22:12:13 GMT</pubDate><content:encoded>&lt;p&gt;Government IT distributor Carahsoft has become the public sector distributor for IonQ’s quantum technology, according to IonQ and Carahsoft. The two companies announced the deal on 9 November 2026. It covers IonQ’s quantum computing, networking, sensing and security offerings. The target buyers are federal agencies, state and local government, the Department of War, civilian agencies, higher education and healthcare organisations.&lt;/p&gt;
&lt;p&gt;Under the agreement, Carahsoft takes on procurement support and pricing, according to IonQ and Carahsoft. It also handles access to applicable federal contract vehicles, plus channel sales and distribution through its reseller partners, for IonQ’s technology. IonQ, for its part, continues building out what it calls a full-stack quantum platform spanning computing, networking, sensing and security.&lt;/p&gt;
&lt;p&gt;Steve Harris, IonQ’s senior vice president and general manager of public sector, said the partnership would make the company’s quantum offerings “more readily accessible” to government customers. Those offerings span computing, networking, security and sensing. He framed the move as coming at a critical period for the sector.&lt;/p&gt;
&lt;p&gt;IonQ ties the deal to Executive Order 14413 on quantum innovation. It points to recent progress it says strengthens its case with government buyers. That includes advancing to Stage C of DARPA’s Quantum Benchmarking Initiative. It also includes a quantum-networking collaboration with the Air Force Research Laboratory in Rome, New York.&lt;/p&gt;
&lt;p&gt;Neither company has stated a dollar figure, a specific contract vehicle or a named pilot agency for the new arrangement.&lt;/p&gt;
</content:encoded><dc:creator>Owen Hartley</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-AFRL_provides_US_with_robust_future_quantum_computing__networking_capabilities_(8164033)_Z1caNjE.jpg" medium="image"/><category>business</category></item><item><title>Which European Quantum Computing Companies Are There, and How Much Have They Raised?</title><link>https://explorequantum.com/article/which-european-quantum-companies-are-there/</link><guid isPermaLink="true">https://explorequantum.com/article/which-european-quantum-companies-are-there/</guid><description>Five European quantum computing companies, five hardware approaches and five funding rounds: IQM, Pasqal, Alice &amp; Bob, Quantum Motion and planqc, with what each builds and what it has raised.</description><pubDate>Mon, 09 Nov 2026 16:27:15 GMT</pubDate><content:encoded>&lt;p&gt;Five companies anchor Europe’s quantum computing industry. Each has picked a different way to build a qubit. IQM Quantum Computers, based in Espoo, Finland, raised $320 million, about €275 million, in a Series B round announced on 3 September 2025. Ten Eleven Ventures led the round, with strong support from Finnish investor Tesi. The round brought IQM’s total funding to $600 million. IQM calls it the largest Series B ever raised in the quantum sector in Europe or outside the United States.&lt;/p&gt;
&lt;h2 id=&quot;iqm-builds-superconducting-chips-for-on-premises-and-cloud-use&quot;&gt;IQM builds superconducting chips for on-premises and cloud use&lt;/h2&gt;
&lt;p&gt;According to IQM, it builds its own superconducting chips and full-stack quantum computers. It sells them for on-premises deployment, alongside cloud platform access, to enterprises, research institutions, universities, HPC centres and national laboratories worldwide. Superconducting quantum circuits are one of the two main hardware approaches research institute QuTech points to for building a qubit, alongside the silicon spin qubits Quantum Motion uses. Both, it says, must overcome the same obstacle: qubits are extremely fragile and perturbed by the slightest fluctuation in their surroundings. The September 2025 Series B gave it the funds to scale that manufacturing and sales effort.&lt;/p&gt;
&lt;h2 id=&quot;pasqals-neutral-atoms-and-a-route-to-a-dual-listing&quot;&gt;Pasqal’s neutral atoms and a route to a dual listing&lt;/h2&gt;
&lt;p&gt;A team including Nobel laureate Alain Aspect founded Pasqal in 2019; the company is headquartered in Palaiseau, France. It builds neutral-atom quantum computers that use laser-trapped rubidium atoms as qubits. According to Wikipedia, that approach encodes each qubit in two energy levels of a single neutral atom held in place by optical tweezers, individual focused laser beams. For a two-qubit gate, nearby atoms are briefly excited to a high-energy Rydberg state, where their interactions let one atom’s state control another’s. It says it has ten quantum computers built, with seven already deployed and three in production. That includes on-site systems delivered to HPC centres in France and Germany. According to Pasqal, it announced on 4 March 2026 a financing of at least €340 million. It called the package a first step toward a dual public listing on Nasdaq and Euronext Paris, through a business combination with a special-purpose acquisition vehicle. The package splits into a €170 million private funding round, from investors including Parkway, Quanta Computer, LG Electronics and CMA CGM, and about €170 million, roughly $200 million, in convertible financing led by Inflection Point. Bpifrance Large Venture is the anchor investor on that convertible piece, and the deal values Pasqal at $2 billion.&lt;/p&gt;
&lt;h2 id=&quot;alice--bob-needs-fewer-physical-qubits-per-error-corrected-qubit&quot;&gt;Alice &amp;amp; Bob needs fewer physical qubits per error-corrected qubit&lt;/h2&gt;
&lt;p&gt;Based in Paris, Alice &amp;amp; Bob was founded in 2020 and builds what it calls “cat qubits”. According to Wikipedia, a cat qubit encodes information not in a two-level system but in superpositions of coherent states of a single oscillator mode, such as a microwave cavity. Its main source of energy loss then shows up as a phase error rather than a direct switch between states, which suppresses bit-flip errors exponentially compared with a conventional qubit. It says the design needs 50 or fewer physical qubits to reach the error protection that a transmon-based logical qubit would otherwise need more than 1,000 physical qubits for. It is aiming for an early fault-tolerant quantum computer with 100 high-fidelity logical qubits by 2030, across chip generations it names Boson, Helium, Lithium, Beryllium and Graphene. According to Alice &amp;amp; Bob, it closed a €100 million Series B round on 28 January 2025. Future French Champions, AXA Venture Partners and Bpifrance led that round, with Series A investors Elaia Partners, Breega and Supernova Invest also taking part. The round brought its total funding to €130 million. In May 2026 it added an extension to that same Series B, with an investment from NVIDIA’s venture arm, NVentures.&lt;/p&gt;
&lt;h2 id=&quot;quantum-motion-puts-silicon-spin-qubits-on-standard-chip-lines&quot;&gt;Quantum Motion puts silicon spin qubits on standard chip lines&lt;/h2&gt;
&lt;p&gt;Quantum Motion spun out of University College London and the University of Oxford in 2017 and builds quantum computers using silicon spin qubits. QuTech describes that approach as electron spins held in quantum dots. It makes them with industry-standard CMOS transistor manufacturing, which it says lets utility-scale systems fit into standard data-centre racks. In 2025 it deployed what it describes as the world’s first full-stack quantum computer on standard silicon chips, at the UK’s National Quantum Computing Centre. According to Quantum Motion, it raised $160 million in a Series C round announced on 7 May 2026. DCVC and Kembara co-led the round, with new participation from the British Business Bank and Firgun. Returning investors included Oxford Science Enterprises, Inkef, Bosch Ventures, Porsche Automobil Holding SE and Parkwalk Advisors. It describes itself as the UK’s best-funded quantum computing company as a result.&lt;/p&gt;
&lt;h2 id=&quot;planqcs-neutral-atoms-need-no-deep-cooling&quot;&gt;planqc’s neutral atoms need no deep cooling&lt;/h2&gt;
&lt;p&gt;Scientists from the Max Planck Institute of Quantum Optics and Ludwig Maximilian University of Munich founded planqc in 2022; it is based in Garching near Munich. It builds quantum computers that use individual neutral atoms, held in optical lattices, crystals of light, as qubits, the same neutral-atom approach Pasqal uses. Wikipedia describes that approach as encoding a qubit in two energy levels of a single trapped atom such as rubidium. While the atoms themselves are cooled to ultracold temperatures, planqc says the rest of the machine runs at room temperature, with no cryogenics required. planqc raised €50 million in a Series A round reported on 8 July 2024. European family office Catron Holding and the DeepTech &amp;amp; Climate Fonds led the round. Bayern Kapital, the Max Planck Foundation, UVC Partners, Speedinvest and private investors also took part, alongside a non-dilutive grant from Germany’s Federal Ministry of Education and Research.&lt;/p&gt;
&lt;h2 id=&quot;what-each-company-builds-and-has-raised&quot;&gt;What each company builds and has raised&lt;/h2&gt;
&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Company&lt;/th&gt;
&lt;th&gt;Country&lt;/th&gt;
&lt;th&gt;Qubit type&lt;/th&gt;
&lt;th&gt;Latest round&lt;/th&gt;
&lt;th&gt;Total raised&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;IQM&lt;/td&gt;
&lt;td&gt;Finland&lt;/td&gt;
&lt;td&gt;Superconducting&lt;/td&gt;
&lt;td&gt;$320 million Series B, Sept 2025.&lt;/td&gt;
&lt;td&gt;$600 million.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Pasqal&lt;/td&gt;
&lt;td&gt;France&lt;/td&gt;
&lt;td&gt;Neutral atom&lt;/td&gt;
&lt;td&gt;At least €340 million financing, March 2026.&lt;/td&gt;
&lt;td&gt;Not stated in this round.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Alice &amp;amp; Bob&lt;/td&gt;
&lt;td&gt;France&lt;/td&gt;
&lt;td&gt;Cat qubits&lt;/td&gt;
&lt;td&gt;€100 million Series B, Jan 2025, plus 2026 extension.&lt;/td&gt;
&lt;td&gt;€130 million before the extension.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Quantum Motion&lt;/td&gt;
&lt;td&gt;United Kingdom&lt;/td&gt;
&lt;td&gt;Silicon spin&lt;/td&gt;
&lt;td&gt;$160 million Series C, May 2026.&lt;/td&gt;
&lt;td&gt;Not stated in this round.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;planqc&lt;/td&gt;
&lt;td&gt;Germany&lt;/td&gt;
&lt;td&gt;Neutral atom (optical lattice)&lt;/td&gt;
&lt;td&gt;€50 million Series A, July 2024.&lt;/td&gt;
&lt;td&gt;Not stated in this round.&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p&gt;Four of these five raised fresh money within the past 18 months, and Pasqal’s March 2026 package is explicitly a step toward becoming a publicly listed company rather than a private one.&lt;/p&gt;
</content:encoded><dc:creator>Owen Hartley</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-Alice_and_Bob_Cryostat_12ccqE.jpg" medium="image"/><category>business</category></item><item><title>Which Quantum Computing Companies Are Publicly Listed?</title><link>https://explorequantum.com/article/which-quantum-computing-companies-are-publicly-listed/</link><guid isPermaLink="true">https://explorequantum.com/article/which-quantum-computing-companies-are-publicly-listed/</guid><description>Three pure-play quantum computing companies trade on US exchanges today: IonQ on the NYSE, Rigetti Computing and D-Wave Quantum on the Nasdaq. Here is what each one does and its most recently reported revenue and losses.</description><pubDate>Mon, 09 Nov 2026 10:11:46 GMT</pubDate><content:encoded>&lt;p&gt;Three pure-play quantum computing companies trade on public exchanges today. According to Wikipedia, IonQ was the first to list, going public in October 2021 through a merger with a special-purpose acquisition company. It raised about $636 million and trades on the New York Stock Exchange under the ticker IONQ. According to Wikipedia, Rigetti Computing and D-Wave Quantum followed in 2022. All three now report quarterly revenue and losses like any other listed company.&lt;/p&gt;
&lt;h2 id=&quot;ionq-nyse-ionq&quot;&gt;IonQ (NYSE: IONQ)&lt;/h2&gt;
&lt;p&gt;IonQ is a trapped-ion quantum computing company based in College Park, Maryland. According to IonQ, its main business is building commercial quantum computers accessed through the cloud. It also sells quantum networking and security, quantum sensing, and quantum space-infrastructure products, on what it calls a full-stack quantum platform. According to IonQ, it reported revenue of $173.8 million for the third quarter of 2026, up 336% from a year earlier. It raised its full-year 2026 revenue guidance to $505-515 million. It posted GAAP net income of $564.0 million, which it says mainly reflects a non-cash accounting gain on warrants rather than operating performance. Its adjusted EBITDA loss was $133.6 million.&lt;/p&gt;
&lt;h2 id=&quot;rigetti-computing-nasdaq-rgti&quot;&gt;Rigetti Computing (Nasdaq: RGTI)&lt;/h2&gt;
&lt;p&gt;Rigetti Computing was founded in 2013 by physicist Chad Rigetti and is based in Berkeley, California. It builds superconducting quantum computers and quantum processors, on what it calls an integrated systems approach. According to Rigetti, it offers cloud access to its machines through its Quantum Cloud Services platform. It also runs its own chip-fabrication facility, Fab-1, in Fremont, California. According to Wikipedia, the company went public in March 2022 through a merger with a special-purpose acquisition company and trades on the Nasdaq. For the second quarter of 2026, Rigetti reported total revenue of $5.1 million, up 185% from a year earlier. It posted an operating loss of $28.1 million and a GAAP net loss of $52.6 million, or a non-GAAP net loss of $16.0 million. It held $541.3 million in cash, cash equivalents and investments as of 30 June 2026, with no debt.&lt;/p&gt;
&lt;h2 id=&quot;d-wave-quantum-nasdaq-qbts&quot;&gt;D-Wave Quantum (Nasdaq: QBTS)&lt;/h2&gt;
&lt;p&gt;According to Wikipedia, D-Wave was founded in 1999 in Burnaby, British Columbia. It develops and sells quantum computing systems and services. According to D-Wave, it is the only company offering both annealing and gate-model quantum computers. Its products include its Advantage2 quantum computer, its Leap cloud service and its Ocean software suite. D-Wave went public in August 2022 through a merger with a special-purpose acquisition company, first listing on the New York Stock Exchange. It moved that listing to the Nasdaq, under the same ticker, on 27 July 2026. For the second quarter of 2026, D-Wave reported revenue of $3.1 million, flat from a year earlier, and a net loss of $48.0 million, or $0.13 per share. It held $546.2 million in consolidated cash and marketable investment securities as of 30 June 2026, which it called its highest-ever cash balance at the time. D-Wave is also expanding by acquisition. According to D-Wave, it announced on 7 January 2026 that it agreed to acquire Quantum Circuits Inc, a gate-model quantum computing company, for $550 million. The deal splits into about $300 million in D-Wave stock and $250 million in cash, to add gate-model hardware alongside its existing annealing systems.&lt;/p&gt;
&lt;h2 id=&quot;the-three-companies-compared&quot;&gt;The three companies compared&lt;/h2&gt;
&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Company&lt;/th&gt;
&lt;th&gt;Ticker&lt;/th&gt;
&lt;th&gt;Listed&lt;/th&gt;
&lt;th&gt;Most recent quarterly revenue&lt;/th&gt;
&lt;th&gt;Most recent quarterly loss or income&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;IonQ&lt;/td&gt;
&lt;td&gt;NYSE: IONQ&lt;/td&gt;
&lt;td&gt;October 2021.&lt;/td&gt;
&lt;td&gt;$173.8 million, Q3 2026.&lt;/td&gt;
&lt;td&gt;GAAP net income $564.0 million, mostly a non-cash warrant gain; adjusted EBITDA loss $133.6 million.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Rigetti Computing&lt;/td&gt;
&lt;td&gt;Nasdaq: RGTI&lt;/td&gt;
&lt;td&gt;March 2022.&lt;/td&gt;
&lt;td&gt;$5.1 million, Q2 2026.&lt;/td&gt;
&lt;td&gt;GAAP net loss $52.6 million.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;D-Wave Quantum&lt;/td&gt;
&lt;td&gt;Nasdaq: QBTS&lt;/td&gt;
&lt;td&gt;August 2022.&lt;/td&gt;
&lt;td&gt;$3.1 million, Q2 2026.&lt;/td&gt;
&lt;td&gt;Net loss $48.0 million.&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p&gt;Rigetti’s and D-Wave’s figures above are their latest published results, from the second quarter of 2026. This page gets refreshed each quarter as new results come out.&lt;/p&gt;
</content:encoded><dc:creator>Owen Hartley</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-Hot_off_the_press_%E2%80%94_the_latest_D-Wave_wafer_of_quantum_processors_and_TIME_cover_story_ZWk0H9.jpg" medium="image"/><category>business</category></item><item><title>What Do Quantum Companies Actually Sell, to Whom, and What Does It Earn Them?</title><link>https://explorequantum.com/article/what-do-quantum-companies-actually-sell/</link><guid isPermaLink="true">https://explorequantum.com/article/what-do-quantum-companies-actually-sell/</guid><description>Quantum companies sell full systems, cloud access and component supply to enterprises, governments and each other: IonQ reported $173.8 million in third-quarter 2026 revenue, IQM €8.9 million, and suppliers such as QuantWare and Qblox sell parts other builders use.</description><pubDate>Mon, 09 Nov 2026 06:26:38 GMT</pubDate><content:encoded>&lt;p&gt;Quantum companies sell three different things today. Those are full computers, access to computers they keep running themselves, and the chips and electronics that go inside someone else’s computer. According to IonQ, the largest of the pure-play quantum companies, it reported record third-quarter 2026 revenue of $173.8 million, up 336% year over year.&lt;/p&gt;
&lt;h2 id=&quot;full-systems-to-enterprises-governments-and-labs&quot;&gt;Full systems to enterprises, governments and labs&lt;/h2&gt;
&lt;p&gt;The best-known layer is the full quantum computer. A customer buys it and installs it or runs it in a dedicated location. According to IQM Quantum Computers, it sells full-stack superconducting quantum computers to enterprises, universities, supercomputing centres and national laboratories. IQM says it has sold 31 systems and delivered 17 of them so far. According to Rigetti, it sells on-premises superconducting systems from 9 to 108 qubits, including its Novera and Cepheus product lines. Its buyers are national laboratories, universities and research organisations. According to Pasqal, it sells neutral-atom quantum computers to enterprise, sovereign and public-sector customers, naming Saudi Aramco, Crédit Agricole and LG Electronics among its buyers.&lt;/p&gt;
&lt;p&gt;These sales come with long tails attached. According to IonQ, its signed contracts still to be delivered, known as remaining performance obligations, more than doubled to $1.2 billion. According to IQM, new order intake reached €42.2 million in the third quarter of 2026. A €33 million contract with Finland’s CSC, for its Halocene error-correction platform, drove most of that. IQM’s order backlog now stands at €102.5 million. According to Pasqal, its booked and awarded business stood at €69.8 million, even though it sold no new processors in the quarter.&lt;/p&gt;
&lt;h2 id=&quot;cloud-access-for-customers-who-never-buy-the-hardware&quot;&gt;Cloud access, for customers who never buy the hardware&lt;/h2&gt;
&lt;p&gt;Many customers never buy a machine at all; they rent time on one. According to IBM, it sells access to its quantum computers under five plans. A free Open Plan gives up to 10 minutes of runtime a month, and a Pay-As-You-Go plan bills by the second. A prepaid Flex Plan starts at 400 minutes a year, and an annual Premium Plan starts at 5,200 minutes a year. An On-Prem Plan puts a dedicated, IBM-run system at the customer’s own site, priced on request. According to IBM, clients signing Flex, Premium or On-Prem contracts worth $250,000 or more join the IBM Quantum Network, its membership programme for industry and research users.&lt;/p&gt;
&lt;p&gt;According to D-Wave, it sells cloud access to its annealing and gate-model systems through its Leap service. It counts more than 100 commercial, government and research customers for that service and its on-premises installations. The split between hardware sales and services shows up in the numbers. According to D-Wave, 67.7% of its first-half 2026 revenue came from commercial customers, up from 16% a year earlier. Total revenue still fell to $5.9 million from $18.1 million. Bookings, the value of new contracts signed rather than revenue already recognised, rose more than elevenfold to $35.5 million in the same period. According to Pasqal, its services revenue, consulting, remote access, upgrades and maintenance for its machines, rose 53% to €3.1 million in the third quarter of 2026.&lt;/p&gt;
&lt;h2 id=&quot;selling-to-other-quantum-companies-not-to-end-users&quot;&gt;Selling to other quantum companies, not to end users&lt;/h2&gt;
&lt;p&gt;A third layer sells neither systems nor access to end customers. It sells the parts a quantum-hardware builder needs to assemble its own machine. According to QuantWare, it sells ready-made quantum processors, chip packaging and foundry services to other companies and institutes building quantum computers, rather than building a computer itself. It says it has shipped chips to more than 50 customers in 20 countries. QuantWare calls itself the largest commercial QPU supplier by volume. It raised a $178 million Series B round in May 2026 to expand that role.&lt;/p&gt;
&lt;p&gt;According to Qblox, it sells control electronics and software that generate and read the microwave pulses steering qubits. That is not a quantum computer itself, but a part sold to companies and research labs building one. Qblox names Oxford Quantum Circuits, Rigetti, IQM, QuantWare and Qilimanjaro among its customers. It says its control stacks run inside deployed systems such as the MareNostrum Ona machine at the Barcelona Supercomputing Center and the Dutch Quantum Inspire system.&lt;/p&gt;
&lt;p&gt;According to IonQ, $83.5 million of its third-quarter 2026 revenue, nearly half, came from foundry services at SkyWater Technology, the chip manufacturer it has acquired. The remaining $90.3 million came from its own quantum computing platform. That segment covers systems, cloud access, consulting and security work sold to enterprises, universities, US government agencies and defence contractors.&lt;/p&gt;
&lt;h2 id=&quot;what-the-numbers-add-up-to&quot;&gt;What the numbers add up to&lt;/h2&gt;
&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Company&lt;/th&gt;
&lt;th&gt;Revenue&lt;/th&gt;
&lt;th&gt;What it sells&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;IonQ&lt;/td&gt;
&lt;td&gt;$173.8 million in Q3 2026.&lt;/td&gt;
&lt;td&gt;Systems, cloud access, consulting, security work; plus foundry services via SkyWater.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;IQM&lt;/td&gt;
&lt;td&gt;€8.9 million in Q3 2026.&lt;/td&gt;
&lt;td&gt;Full-stack superconducting systems to enterprises, universities, supercomputing centres, national labs.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Rigetti&lt;/td&gt;
&lt;td&gt;$9.5 million in H1 2026.&lt;/td&gt;
&lt;td&gt;On-premises systems and cloud access via Quantum Cloud Services.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;D-Wave&lt;/td&gt;
&lt;td&gt;$5.9 million in H1 2026.&lt;/td&gt;
&lt;td&gt;On-premises systems and cloud access via Leap.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Pasqal&lt;/td&gt;
&lt;td&gt;€3.3 million in Q3 2026.&lt;/td&gt;
&lt;td&gt;Neutral-atom systems, consulting, remote access, upgrades and maintenance.&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p&gt;None of the pure-play hardware makers yet earns what the large cloud providers or chipmakers do. Their revenue sits in the single-digit to low-hundred-million range per quarter. Their backlogs and bookings often run several times that size.&lt;/p&gt;
</content:encoded><dc:creator>Owen Hartley</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-The_Beauty_of_Quantum_Computers_Z2l44jD.jpg" medium="image"/><category>business</category></item><item><title>HKMA and HKUST Launch Post-Quantum Cryptography Toolkit for Hong Kong&apos;s Banks</title><link>https://explorequantum.com/article/hkma-hkust-pqc-toolkit-banks/</link><guid isPermaLink="true">https://explorequantum.com/article/hkma-hkust-pqc-toolkit-banks/</guid><description>The Hong Kong Monetary Authority and HKUST have launched a toolkit to help the city&apos;s banks plan their move to post-quantum cryptography, as the sector&apos;s readiness score sits at 2.3 out of 10.</description><pubDate>Mon, 09 Nov 2026 02:06:56 GMT</pubDate><content:encoded>&lt;p&gt;The Hong Kong Monetary Authority (HKMA) and the Hong Kong University of Science and Technology (HKUST) have launched a Post-Quantum Cryptography (PQC) Toolkit for the city’s licensed banks, according to HKMA and HKUST. HKMA issued a circular on the toolkit on 2 November. It also publicised the launch at its FiNETech8 forum, with industry participants supporting the toolkit’s development. The goal is to turn banks’ awareness of quantum risk into concrete transition plans, ahead of the sector’s move to quantum-safe cryptography.&lt;/p&gt;
&lt;p&gt;The launch comes against a low starting score. HKMA tracks bank readiness with its Quantum Preparedness Index. That index stood at 2.3 out of 10 in a baseline assessment from July 2026. The authority wants the toolkit to help banks lift that score to a fully quantum-ready 10 out of 10 by 2030.&lt;/p&gt;
&lt;p&gt;The toolkit covers four practical steps, according to HKMA and HKUST. First, a cryptographic inventory of where cryptography is used across a bank’s core systems. Second, a map of which systems depend on it, and which to migrate first. Third, governance with clear executive and board accountability. Fourth, engaging suppliers, such as cloud providers and hardware-security-module vendors, on their own quantum-safe plans.&lt;/p&gt;
&lt;p&gt;According to HKMA and HKUST, HKMA deputy chief executive Arthur Yuen said the toolkit gives banks a common foundation they can adapt to their own circumstances. He said that should help them turn quantum-risk awareness into clear responsibilities and plans. HKUST Business School dean Frederik Anseel described preparing for the quantum era as a leadership challenge for banks’ boards and executives, not just a technical one.&lt;/p&gt;
&lt;p&gt;The next milestone is 2030, the date by which HKMA wants the sector’s Quantum Preparedness Index to reach 10 out of 10.&lt;/p&gt;
</content:encoded><dc:creator>Owen Hartley</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-Hong_Kong_Monetary_Authority_ZkrT8B.jpg" medium="image"/><category>security</category></item><item><title>What Is PsiQuantum Building, Where, and When Will It Be Ready?</title><link>https://explorequantum.com/article/what-is-psiquantum-building-and-when-will-it-be-ready/</link><guid isPermaLink="true">https://explorequantum.com/article/what-is-psiquantum-building-and-when-will-it-be-ready/</guid><description>PsiQuantum is building two utility-scale photonic quantum computers, one near Brisbane and one in Chicago, with construction under way at both sites and CEO Jeremy O&apos;Brien targeting an error-corrected machine by 2029.</description><pubDate>Sun, 08 Nov 2026 18:09:24 GMT</pubDate><content:encoded>&lt;p&gt;PsiQuantum chief executive Jeremy O’Brien has named 2029 as the target year for an error-corrected quantum computer. He set that date in April 2024 and later reaffirmed it. It stands apart from the construction milestones at the company’s two building sites, in Australia and the United States. PsiQuantum no longer ties either site to a fixed completion date for the full machine.&lt;/p&gt;
&lt;p&gt;PsiQuantum builds quantum computers on photonic qubits. These are units of quantum information carried by single particles of light, rather than by superconducting circuits or trapped atoms. A photonic qubit stores information in a property of that particle, such as its polarisation or which of two paths it travels. An optical element called a beam splitter can place the photon into a superposition of two paths or states, which forms the core of the qubit. PsiQuantum says this lets it build its hardware in standard semiconductor fabs, using the precision and volume those fabs already have, instead of needing exotic materials. According to PsiQuantum, it demonstrated that manufacturable approach on 26 February 2025 with a chipset called Omega, made at GlobalFoundries in upstate New York on the same process equipment used for ordinary chips. Measured fidelities ranged from 99.22% for a two-qubit fusion gate to 99.98% for single-qubit state preparation and readout.&lt;/p&gt;
&lt;p&gt;PsiQuantum is building two full-scale machines, each aimed at roughly 1 million physical qubits. In Australia, the federal and Queensland governments pledged a combined 940 million Australian dollars, then about 620 million US dollars, on 29 April 2024. The money was for PsiQuantum to set up its Asia-Pacific headquarters there. It was also to build what both governments called the world’s first utility-scale, fault-tolerant quantum computer, with the end of 2027 named then as the target completion year. According to PsiQuantum, that machine was first planned for a site near Brisbane Airport, but no construction had started there. In 2026 PsiQuantum moved the project to Moreton Bay Central in Petrie, Queensland, and broke ground on 18 June 2026. According to PsiQuantum, the Moreton Bay site will house tens of thousands of photonic chips and large-scale cryogenic infrastructure. The cryogenic plant there, supplied by Linde Engineering and described as one of the largest ever built for quantum computing, should be finished in the second half of 2027. It no longer gives a separate target date for the complete, working machine at that site.&lt;/p&gt;
&lt;p&gt;In the United States, PsiQuantum announced on 25 July 2024 that it would anchor the country’s first utility-scale, fault-tolerant quantum computer. The site is the former United States Steel South Works steel mill in Chicago. It became the Illinois Quantum and Microelectronics Park, funded by the state of Illinois and DARPA alongside the company itself. PsiQuantum broke ground on its plot there on 30 September 2025, calling it the start of construction on “America’s first million-qubit scale, fault-tolerant quantum computer”. It gives no year for the full machine’s completion; the wider park’s developer has put the first construction phase at 2027.&lt;/p&gt;
&lt;p&gt;Construction is under way on both continents. Cryogenic infrastructure at Moreton Bay and the first building phase in Chicago are both due in 2027. O’Brien’s 2029 target for an error-corrected machine remains the company’s own answer to when the technology itself, not the buildings around it, should work.&lt;/p&gt;
</content:encoded><dc:creator>Mara Quinlan</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-Aerial_photograph_of_Globalfoundries_Dresden_Z1Gi9J9.jpg" medium="image"/><category>computing</category></item><item><title>What Is Microsoft&apos;s Majorana Qubit, and What Has It Shown?</title><link>https://explorequantum.com/article/what-is-microsofts-majorana-qubit/</link><guid isPermaLink="true">https://explorequantum.com/article/what-is-microsofts-majorana-qubit/</guid><description>Microsoft&apos;s Majorana qubit tries to store quantum information in the shape of exotic quasiparticles rather than a single particle&apos;s state. The company has unveiled two chips, but its own Nature paper stops short of confirming the particles are there.</description><pubDate>Sun, 08 Nov 2026 16:36:44 GMT</pubDate><content:encoded>&lt;p&gt;A Majorana qubit is Microsoft’s attempt to store quantum information in a quasiparticle pair instead of a single particle’s fragile state. It is the design behind the two chips Microsoft has unveiled so far, Majorana 1 and Majorana 2, the latest in June 2026.&lt;/p&gt;
&lt;h2 id=&quot;how-the-idea-works&quot;&gt;How the idea works&lt;/h2&gt;
&lt;p&gt;The underlying design is a topological qubit. The approach was theorised in 1997 by physicists Alexei Kitaev and Michael Freedman. It encodes information in the overall shape, or “braiding”, of exotic quasiparticles called anyons, where no single particle’s state carries it. Because the stored information depends on topology, not on fine local detail, small local disturbances should not corrupt it as easily as they do other qubit designs.&lt;/p&gt;
&lt;p&gt;Microsoft’s version relies on Majorana zero modes. These are quasiparticles predicted to form at the ends of certain superconducting nanowires, each its own antiparticle. A pair of them is expected to store one bit of quantum information jointly, split between both ends and not held in either one alone. That shared storage is the whole appeal: disturb one end, and the stored bit should still be recoverable from the pair.&lt;/p&gt;
&lt;p&gt;The qubit is named after Ettore Majorana. He was an Italian theoretical physicist who in 1937 worked out a theory in which neutral particles such as neutrinos could be their own antiparticle, a class now called Majorana fermions. Majorana disappeared on 25 March 1938, aged 31. A Rome court closed that case in January 2025 by declaring him presumed dead.&lt;/p&gt;
&lt;h2 id=&quot;what-microsoft-has-unveiled&quot;&gt;What Microsoft has unveiled&lt;/h2&gt;
&lt;p&gt;According to Microsoft, it unveiled its first chip, Majorana 1, on 19 February 2025, calling it the first quantum processing unit built on a “topological core”. The eight-qubit chip is made from a material Microsoft calls a topoconductor, a hybrid of indium arsenide and aluminium, which the company says can host Majorana zero modes. Microsoft says the same architecture can eventually scale to a million qubits on one chip.&lt;/p&gt;
&lt;p&gt;Alongside that unveiling, Microsoft said it had been selected as one of two companies to advance to the final phase of DARPA’s Underexplored Systems for Utility-Scale Quantum Computing programme. Under that programme it aims to build a fault-tolerant prototype topological quantum computer “in years, not decades”.&lt;/p&gt;
&lt;p&gt;Microsoft’s claim that Majorana 1 holds working topological qubits goes further than its own companion paper in Nature actually states, published the same day as the unveiling. That paper describes a new way to read out the parity of a hybrid semiconductor-superconductor device. But it says explicitly that these measurements “do not, by themselves, determine whether the low-energy states detected by interferometry are topological”. In other words, the published science supports a narrower claim than the press announcement.&lt;/p&gt;
&lt;p&gt;A gap like that matters because Microsoft-affiliated researchers made a similar claim earlier and had to withdraw it. Researchers at TU Delft’s QuTech institute published a 2018 Nature paper claiming conclusive evidence for Majorana zero modes. A TU Delft Research Integrity Committee investigation found the researchers had selected data that supported the phenomenon they were looking for, while omitting data that could have raised questions about the conclusions. The authors retracted the paper on 8 March 2021.&lt;/p&gt;
&lt;h2 id=&quot;majorana-2-and-the-reliability-claim&quot;&gt;Majorana 2 and the reliability claim&lt;/h2&gt;
&lt;p&gt;Microsoft unveiled its second-generation chip, Majorana 2, on 2 June 2026 and says it is 1,000 times more reliable than Majorana 1. It reports qubit lifetimes exceeding 20 seconds, occasionally over one minute, up from 1 to 12 milliseconds before. That gain follows swapping Majorana 1’s aluminium superconductor for lead. Microsoft also says it has cut its timeline in half, now targeting 2029 for a scalable, practical quantum computer. It calls that a milestone on the path to a fault-tolerant machine, not a fault-tolerant machine itself.&lt;/p&gt;
&lt;p&gt;Those figures are Microsoft’s own characterisation of its hardware. The 2029 target is the next date to watch, on Microsoft’s own revised timeline.&lt;/p&gt;
</content:encoded><dc:creator>Mara Quinlan</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-Building_92_of_Microsoft_Redmond_Campus_-_panoramio_LRimr.jpg" medium="image"/><category>computing</category></item><item><title>What Can Quantinuum&apos;s Helios Quantum Computer Do, and How Does It Compare With Earlier Systems?</title><link>https://explorequantum.com/article/what-can-quantinuums-helios-do/</link><guid isPermaLink="true">https://explorequantum.com/article/what-can-quantinuums-helios-do/</guid><description>Quantinuum&apos;s Helios packs 98 trapped-ion qubits, up to 48 error-corrected logical qubits and 99.921% two-qubit fidelity. This explainer answers what it can do today, how it compares with its predecessor H2, and what comes next on Quantinuum&apos;s roadmap.</description><pubDate>Sun, 08 Nov 2026 10:26:07 GMT</pubDate><content:encoded>&lt;p&gt;Quantinuum put a 98-qubit computer called Helios on commercial sale on 5 November 2025, according to the company. Helios is a general-purpose trapped-ion machine built on a QCCD (quantum charge-coupled device) architecture, in which qubits move through a ring-shaped storage and computation network between memory, cache and logic zones. That mobility lets any qubit entangle with any other, not just its neighbours. Quantinuum reports a two-qubit gate fidelity of 99.921% across all qubit pairs and a one-qubit gate fidelity of 99.9975%. The company calls it the most accurate commercial quantum computer available. According to Quantinuum, those 98 physical qubits can also be configured into up to 48 fully error-corrected logical qubits, with 99.99% state preparation fidelity. Up to 50 logical qubits outperform physical qubits in a magnetism simulation, and up to 94 logical qubits, error-detected rather than fully error-corrected, can be globally entangled with better-than-physical performance.&lt;/p&gt;
&lt;h2 id=&quot;how-the-hardware-works&quot;&gt;How the hardware works&lt;/h2&gt;
&lt;p&gt;Helios is a trapped-ion machine: electromagnetic fields hold charged atoms in place. Each ion’s stable electronic states store a qubit, and lasers manipulate them. Entanglement arises through the ions’ shared motion. Helios’s ring-shaped network is a QCCD (quantum charge-coupled device) architecture. It splits the trap into separate zones with electrodes, so ions can shuttle between storage and interaction zones instead of all sitting in one trap. That mobility is what lets any qubit pair up with any other.&lt;/p&gt;
&lt;p&gt;The “logical qubit” figures above describe error-protected qubits. Quantum information is encoded across several entangled physical qubits rather than stored in just one. Individual physical qubits stay prone to errors from decoherence and environmental noise. Error correction reads the collective “syndrome” of the physical qubits. That catches and fixes mistakes without disturbing the logical information itself.&lt;/p&gt;
&lt;h2 id=&quot;helios-versus-system-model-h2&quot;&gt;Helios versus System Model H2&lt;/h2&gt;
&lt;p&gt;Helios succeeds Quantinuum’s System Model H2. H2 had 56 fully connected qubits on a racetrack-shaped QCCD architecture and a two-qubit gate fidelity above 99.9%. Helios nearly doubles the qubit count to 98, keeps the full connectivity, and reaches a higher two-qubit fidelity of 99.921% across all qubit pairs, according to Quantinuum.&lt;/p&gt;
&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;&lt;/th&gt;
&lt;th&gt;System Model H2&lt;/th&gt;
&lt;th&gt;Helios&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Physical qubits&lt;/td&gt;
&lt;td&gt;56.&lt;/td&gt;
&lt;td&gt;98.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Connectivity&lt;/td&gt;
&lt;td&gt;Fully connected.&lt;/td&gt;
&lt;td&gt;Fully connected.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Architecture&lt;/td&gt;
&lt;td&gt;Racetrack QCCD.&lt;/td&gt;
&lt;td&gt;Ring-shaped QCCD network.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Two-qubit gate fidelity&lt;/td&gt;
&lt;td&gt;Above 99.9%.&lt;/td&gt;
&lt;td&gt;99.921%.&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p&gt;From its 98 physical qubits, Helios also produces up to 48 fully error-corrected logical qubits.&lt;/p&gt;
&lt;h2 id=&quot;who-is-using-it-and-where&quot;&gt;Who is using it, and where&lt;/h2&gt;
&lt;p&gt;Since launch, Helios has been available through Quantinuum’s cloud service and as an on-premise installation, the company says. It is built for general-purpose use rather than one application area. According to Quantinuum, it named Amgen, BMW Group, JPMorganChase, SoftBank and the platform BlueQubit as customers already using Helios at launch, for work including materials science and finance. Helios was separately used for research in high-temperature superconductivity at a scale that significantly challenges classical methods. Quantinuum also signed a partnership with Singapore’s National Quantum Office and National Quantum Computing Hub at the Helios launch. It is setting up an R&amp;amp;D and operations centre there, and, per its Q3 2026 results, has Helios assembly underway in Singapore ahead of commissioning in spring 2027.&lt;/p&gt;
&lt;h2 id=&quot;what-comes-next&quot;&gt;What comes next&lt;/h2&gt;
&lt;p&gt;Helios is the first of three machines on Quantinuum’s published roadmap to fault-tolerant quantum computing. The company targets a successor called Sol for the second half of 2027, aiming for 100 logical qubits at a target logical error rate of 10⁻⁵. It plans a universal, fully fault-tolerant system called Apollo for 2029. Those are company targets, not delivered machines.&lt;/p&gt;
</content:encoded><dc:creator>Mara Quinlan</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-Quantum_Computing__Ion_Trapping_(5941055642)_K6Xcw.jpg" medium="image"/><category>computing</category></item><item><title>Riverlane&apos;s Data-Driven Error Model Cuts Surface-Code Logical Errors by Up to 75%</title><link>https://explorequantum.com/article/riverlane-dem-augmentation-cuts-logical-errors/</link><guid isPermaLink="true">https://explorequantum.com/article/riverlane-dem-augmentation-cuts-logical-errors/</guid><description>Riverlane says a data-driven way of enriching the error models behind surface-code decoders cut the logical error rate by about 75% in simulation and by about 14% on real Google hardware data, according to a preprint posted this month.</description><pubDate>Sun, 08 Nov 2026 06:14:36 GMT</pubDate><content:encoded>&lt;p&gt;According to Riverlane’s preprint, a new way of building the error models behind surface-code decoders cut the simulated logical error rate by about 75%. At code distance 15, it lifted the error-suppression factor from roughly 1.71 to roughly 2.06.&lt;/p&gt;
&lt;p&gt;The surface code is the error-correction scheme most superconducting quantum computers plan to use. It turns noisy physical qubits into one reliable logical qubit, and its decoder is only as good as the error model it is given. According to Riverlane’s preprint, the company built a data-driven way to improve that model. Instead of assuming only the error types a decoder designer expects, it mines the decoder’s own syndrome data for correlated error mechanisms a standard model misses, then folds them back in.&lt;/p&gt;
&lt;p&gt;A second test, according to Riverlane’s preprint, used real hardware data from a distance-7 surface-code run on a Google Quantum AI processor. There the gain was smaller: about a 14% drop in the logical error rate and roughly a 9% improvement in the error-suppression factor.&lt;/p&gt;
&lt;p&gt;According to Riverlane’s preprint, the technique combines pairwise error mechanisms from a baseline detector error model. It uncovers higher-order correlated mechanisms, then reshapes the result so graph-based decoders such as PyMatching can still use it. Four researchers are named as authors in Riverlane’s preprint: Zohar Schwartzman-Nowik, Matthew Stafford, Boris Varbanov and Joan Camps. It frames the result as the first data-driven augmentation method of this kind that keeps compatibility with correlated graph-based decoders.&lt;/p&gt;
</content:encoded><dc:creator>Mara Quinlan</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-Google_Sycamore_Chip_001_2q9yYu.jpg" medium="image"/><category>computing</category></item><item><title>UK&apos;s FCA Sets Up World&apos;s First Regulator-Led Quantum Experimentation Group</title><link>https://explorequantum.com/article/uk-fca-quantum-security/</link><guid isPermaLink="true">https://explorequantum.com/article/uk-fca-quantum-security/</guid><description>The Financial Conduct Authority is building a Quantum Experimentation Group to test whether quantum machine learning catches fraud and money laundering better than classical tools, with registration open until 31 December 2026 and work starting in Q1 2027.</description><pubDate>Sun, 08 Nov 2026 02:11:34 GMT</pubDate><content:encoded>&lt;p&gt;The UK’s Financial Conduct Authority is setting up a Quantum Experimentation Group, according to the FCA. It wants to find out whether quantum machine learning can catch fraud and money laundering better than the classical methods banks use today. The regulator built a synthetic dataset itself for the tests, so participants can benchmark their algorithms without touching real customer data.&lt;/p&gt;
&lt;p&gt;Firms, quantum-technology providers and research labs that want to join have until 31 December 2026 to register their interest, according to the FCA. The group itself is due to start work in the first quarter of 2027.&lt;/p&gt;
&lt;p&gt;The regulator wants to pool financial-services firms, quantum-technology providers and national research laboratories, according to the FCA. The idea is that they share what they learn testing quantum approaches, rather than each exploring the field alone. The regulator says the goal is to help firms work out where quantum computing could make a real difference before it sees wider use.&lt;/p&gt;
&lt;p&gt;The regulator describes the group as the first of its kind run by a regulator anywhere, according to the FCA. Jessica Rusu, the authority’s chief data, information and intelligence officer, said the FCA wants to help firms explore where quantum computing could make a meaningful difference. She said early experimentation is essential.&lt;/p&gt;
&lt;p&gt;Firms that want to take part must register before the end of the year. The Quantum Experimentation Group is due to start work in the first quarter of 2027.&lt;/p&gt;
</content:encoded><dc:creator>Owen Hartley</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-12_Endeavour_Square__Stratford_(2026-04-15)_ZOnEjH.jpg" medium="image"/><category>security</category></item><item><title>Scottish Widows Routes UK Workplace Pensions Into Oxford Quantum Circuits and Quantum Motion</title><link>https://explorequantum.com/article/uk-pension-quantum-investment/</link><guid isPermaLink="true">https://explorequantum.com/article/uk-pension-quantum-investment/</guid><description>Scottish Widows is channelling workplace pension savings into Oxford Quantum Circuits and Quantum Motion through an Aberdeen Investments venture growth mandate inside a long-term asset fund, alongside stakes in Nyobolt and Genomics Limited.</description><pubDate>Sat, 07 Nov 2026 22:19:49 GMT</pubDate><content:encoded>&lt;p&gt;Scottish Widows is putting workplace pension savings behind two UK quantum hardware developers, according to IP Group. Lloyds Banking Group’s pension arm has handed a venture growth mandate to Aberdeen Investments. Aberdeen has named Oxford Quantum Circuits and Quantum Motion among its initial investments. The money moves through the CG Scottish Widows Growth long-term asset fund, a structure built so pension schemes can hold stakes in illiquid private companies. IP Group runs the venture growth portfolio inside that fund.&lt;/p&gt;
&lt;p&gt;Oxford Quantum Circuits builds superconducting-qubit quantum computers it sells commercially, according to Scottish Widows and Aberdeen Investments. Quantum Motion is developing silicon spin-qubit processors. Both are early-stage UK hardware firms. They are the kind of pre-commercial business this pension route is meant to help bridge toward the market.&lt;/p&gt;
&lt;p&gt;The quantum hardware stakes sit alongside two other initial commitments from the same mandate, according to IP Group. They are battery technology developer Nyobolt and AI-enabled genomics and healthcare company Genomics Limited. That puts the quantum bets inside a broader push by Scottish Widows and Aberdeen Investments into British deep-tech and life-sciences startups.&lt;/p&gt;
&lt;p&gt;Scottish Widows chief investment officer Peter Fitzgerald said the mandate connects workplace pension customers’ long-term savings with some of Britain’s most innovative, high-growth companies, according to Scottish Widows. He framed the long-term asset fund route as a way to improve retirement outcomes. It also gives UK firms access to follow-on funding as they prepare to commercialise their technology. Nalaka De Silva, Aberdeen Investments’ head of private markets solutions, said the UK has no shortage of scientific excellence, entrepreneurship or talent, according to Aberdeen Investments. What it has lacked, he said, is enough patient domestic capital to help competitive companies expand and grow. Aberdeen Investments manages the broader growth sub-fund for Scottish Widows and appointed IP Group to run the venture growth portfolio where the quantum and other deep-tech stakes sit.&lt;/p&gt;
&lt;p&gt;Scottish Widows, part of Lloyds Banking Group, serves more than 6 million customers. It administers over £300 billion in assets. The company cites that scale to frame how much pension capital the new mandate could eventually direct toward UK quantum and deep-tech firms.&lt;/p&gt;
</content:encoded><dc:creator>Owen Hartley</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-ScottishWidowsHQ_ZRobpj.jpg" medium="image"/><category>business</category></item><item><title>What Did Google Show With The Willow Chip, And What Is The Next Step?</title><link>https://explorequantum.com/article/google-willow-and-beyond/</link><guid isPermaLink="true">https://explorequantum.com/article/google-willow-and-beyond/</guid><description>Google&apos;s Willow chip has produced two separate milestones since December 2024: error correction that improves as it scales, and a verifiable quantum speed-up on a real physics problem. This explainer sets out what each result means and the logical qubit Google is chasing next.</description><pubDate>Sat, 07 Nov 2026 18:11:44 GMT</pubDate><content:encoded>&lt;p&gt;Google’s clearest claim to a verifiable quantum advantage so far came on 22 October 2025. It said its Willow chip had run a new algorithm, Quantum Echoes, and finished 13,000 times quicker than the top classical method managed on a leading supercomputer. The results were published in Nature. “Verifiable” means another quantum computer of similar capability can repeat the calculation and check it. Google’s earlier speed claims could not offer that.&lt;/p&gt;
&lt;p&gt;Willow itself is a 105-qubit chip that Google presented on 9 December 2024. It was built in the company’s own chip facility in Santa Barbara. That is almost double the 53 qubits of Sycamore, the chip Google used in 2019 to claim the first task completed faster than a classical supercomputer. Sycamore finished that task in 200 seconds, against an estimated 10,000 years for the fastest supercomputer of the time. At its unveiling, Google called Willow “state-of-the-art” on a set of benchmarks.&lt;/p&gt;
&lt;p&gt;The chip’s first headline result was not speed but error correction. Google says Willow showed, for the first time, error correction “below threshold”. As it scaled an array of encoded qubits from 3x3 through 5x5 to 7x7, the error rate kept falling, roughly halving at each step. Earlier chips saw the opposite: errors rising as qubits were added. Willow’s T1 time, how long a qubit holds its quantum state, came close to 100 microseconds. That is about five times longer than the previous generation of Google’s chips. Google says that because the error rate kept dropping as more qubits were added, the resulting encoded, or “logical”, qubit stayed stable for longer than the individual physical qubits it was built from. Google calls that result “beyond breakeven” and its most convincing prototype yet for a scalable logical qubit. A logical qubit groups several error-prone physical qubits so they catch and correct each other’s mistakes. That is why a chip needs many physical qubits to produce a handful of reliable logical ones.&lt;/p&gt;
&lt;p&gt;Willow’s second result, also announced at its unveiling, was a speed benchmark. Google says that on Random Circuit Sampling, a standard quantum-computing stress test, the chip completed a task in under five minutes. Google calculated that one of the fastest supercomputers of the time would take 10 septillion (10^25) years to do the same task. For Quantum Echoes, Google worked with the University of California, Berkeley. The two tested the algorithm on two molecules, with 15 and 28 atoms, and compared the result against classical nuclear magnetic resonance (NMR) measurements. The quantum results matched the NMR data. They also surfaced structural detail that standard NMR does not always reveal.&lt;/p&gt;
&lt;p&gt;Google now describes its next step as “Milestone 3” on its own hardware roadmap. The goal is a long-lived logical qubit, built from multiple physical qubits, that stays stable for far longer than any single qubit does today. Willow’s below-threshold result already showed that scaling up can make errors rarer rather than more common. Google says Milestone 3 is meant to turn that into a building block stable enough to move toward a large-scale, fault-tolerant quantum computer for practical use.&lt;/p&gt;
</content:encoded><dc:creator>Mara Quinlan</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-Google_Sycamore_Chip_001_2q9yYu.jpg" medium="image"/><category>computing</category></item><item><title>What Did IBM Promise On Its Quantum Roadmap Since 2020, And What Has It Delivered?</title><link>https://explorequantum.com/article/ibm-roadmap-vs-delivery/</link><guid isPermaLink="true">https://explorequantum.com/article/ibm-roadmap-vs-delivery/</guid><description>IBM has published a public quantum roadmap since 2020 and hit every qubit-count target on it. This explainer lines up each promise against its delivery date, through the November 2025 update that pushed fault tolerance to 2029 and 2033.</description><pubDate>Sat, 07 Nov 2026 16:11:55 GMT</pubDate><content:encoded>&lt;p&gt;IBM has kept every qubit-count promise on its public quantum roadmap since 2020. The one date it has pushed back, according to IBM, is fault tolerance itself. IBM published that first roadmap in September 2020, after the smaller Falcon (27 qubits) and Hummingbird (65 qubits) chips. It named the 127-qubit Eagle chip for 2021, the 433-qubit Osprey chip for 2022, and the 1,121-qubit Condor chip for the end of 2023. At the time, IBM called a fault-tolerant quantum computer “an achievable goal within the coming decade”.&lt;/p&gt;
&lt;p&gt;IBM delivered Eagle on schedule, on 16 November 2021. It was IBM’s first chip to pass 100 qubits. A new packaging technique, using through-silicon vias and multi-level wiring, packed the 127 qubits close enough together. Eagle first went to selected members of the IBM Quantum Network.&lt;/p&gt;
&lt;p&gt;Osprey followed, according to IBM, on 9 November 2022, with 433 qubits, more than three times Eagle’s count. At the same event IBM unveiled IBM Quantum System Two. That modular computer was meant to house Osprey, Condor and their successors, and should be running by the end of 2023, IBM said.&lt;/p&gt;
&lt;p&gt;Condor itself arrived, according to IBM, on 4 December 2023, at 1,121 qubits exactly as promised in 2020. But IBM used that same moment to pivot away from adding more qubits. The newer Heron chip, at 133 qubits, became the basis of IBM’s hardware from then on, not Condor. IBM said Heron performed three to five times better than Eagle. The first IBM Quantum System Two, installed in Yorktown Heights, New York, started out running on three Heron chips rather than on Condor.&lt;/p&gt;
&lt;p&gt;In that December 2023 update, IBM also extended its roadmap to 2033. For the first time, it named a date for fault tolerance: Starling, targeted for 2029. Starling is meant to run 200 &lt;strong&gt;logical qubits&lt;/strong&gt; and 100 million quantum gates, using error correction based on the so-called Gross code. A logical qubit is built by encoding information across several physical qubits, so the resulting unit can absorb errors from the underlying hardware. A 2029 target is more concrete than IBM’s 2020 language about fault tolerance being achievable “within the coming decade”. But it also lands later in that same decade.&lt;/p&gt;
&lt;p&gt;IBM’s most recent update, on 12 November 2025, reported two further processors on the way to Starling. Nighthawk, at 120 qubits, is meant to show quantum advantage by the end of 2026. The experimental Loon chip demonstrated, for the first time, every hardware component error correction needs. That included error-correction decoding within 480 nanoseconds, a year ahead of IBM’s own schedule. IBM also moved chip production to a new 300mm wafer fab in Albany, New York, which it said doubles its development speed.&lt;/p&gt;
&lt;p&gt;That same November 2025 roadmap, according to IBM, sets out the targets through 2033. Nighthawk should show quantum advantage by the end of 2026, with circuits of around 7,500 gates. Starling should compute fault-tolerantly with 200 logical qubits by 2029. The successor Blue Jay should reach 2,000 logical qubits and 1 billion gates by 2033. A &lt;strong&gt;fault-tolerant quantum computer&lt;/strong&gt; is one that automatically corrects errors in its qubits through error-correction codes, rather than merely detecting them.&lt;/p&gt;
&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Chip or system&lt;/th&gt;
&lt;th&gt;Promised&lt;/th&gt;
&lt;th&gt;Delivered&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Eagle&lt;/td&gt;
&lt;td&gt;127 qubits, 2021.&lt;/td&gt;
&lt;td&gt;127 qubits, 16 Nov 2021.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Osprey&lt;/td&gt;
&lt;td&gt;433 qubits, 2022.&lt;/td&gt;
&lt;td&gt;433 qubits, 9 Nov 2022.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Condor&lt;/td&gt;
&lt;td&gt;1,121 qubits, end of 2023.&lt;/td&gt;
&lt;td&gt;1,121 qubits, 4 Dec 2023, then superseded by Heron.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Starling (fault-tolerant)&lt;/td&gt;
&lt;td&gt;200 logical qubits, 2029.&lt;/td&gt;
&lt;td&gt;Not yet due.&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Blue Jay&lt;/td&gt;
&lt;td&gt;2,000 logical qubits, 2033.&lt;/td&gt;
&lt;td&gt;Not yet due.&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;
&lt;p&gt;IBM named a specific number and year for every qubit-count chip, and hit both. The one date it has moved is fault tolerance itself. That moved from a vague decade-long outlook in 2020 to a named 2029 target, held since December 2023. The next scheduled test is Nighthawk’s quantum-advantage claim, due by the end of 2026.&lt;/p&gt;
</content:encoded><dc:creator>Mara Quinlan</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-IBM_Q_system_(Fraunhofer_2)_1yqLhy.jpg" medium="image"/><category>computing</category></item><item><title>Which Quantum Computers Is EuroHPC Installing in Europe, and Who Can Use Them?</title><link>https://explorequantum.com/article/eurohpc-quantum-computers/</link><guid isPermaLink="true">https://explorequantum.com/article/eurohpc-quantum-computers/</guid><description>EuroHPC JU has inaugurated six quantum computers across Poland, Czechia, Germany, France, Spain and Italy since picking the sites in 2022, opened free pilot access to four of them in 2026, and has two more systems, in Luxembourg and the Netherlands, in procurement.</description><pubDate>Sat, 07 Nov 2026 10:17:16 GMT</pubDate><content:encoded>&lt;p&gt;EuroHPC JU has now inaugurated all six quantum computers it picked sites for in 2022. According to EuroHPC JU, it opened free pilot access to four of them. The six sit in Poland, Czechia, Germany, France, Spain and Italy. Each is coupled to an existing EuroHPC supercomputer. According to EuroHPC JU, two more systems are in procurement, for Luxembourg and the Netherlands.&lt;/p&gt;
&lt;h2 id=&quot;how-the-sites-were-chosen&quot;&gt;How the sites were chosen&lt;/h2&gt;
&lt;p&gt;According to EuroHPC JU, it selected the six host sites on 4 October 2022. It planned to co-fund up to 50% of a total investment above EUR 100 million, from the Digital Europe Programme and the member states involved. The hosts were IT4Innovations in Czechia, LRZ in Germany, BSC in Spain, GENCI/CEA in France, CINECA in Italy and PSNC in Poland. Each quantum computer was meant to be coupled to the supercomputer its host already runs, so researchers can combine classical and quantum hardware in one job.&lt;/p&gt;
&lt;h2 id=&quot;the-six-systems-now-all-running&quot;&gt;The six systems, now all running&lt;/h2&gt;
&lt;p&gt;PIAST-Q, according to EuroHPC JU, was the first of the six to open. It is a 20-qubit trapped-ion system in Poznań, built by Alpine Quantum Technologies and coupled to the ALTAIR supercomputer at PSNC. It cost EUR 12.28 million, split equally between EuroHPC JU and Poland’s Ministry of Digital Affairs.&lt;/p&gt;
&lt;p&gt;VLQ, according to EuroHPC JU, followed in Ostrava, a 24-qubit superconducting system built by IQM for a thirteen-partner consortium from eight countries. It is coupled to the Karolina supercomputer at IT4Innovations and cost about EUR 5 million, again split evenly with EuroHPC JU.&lt;/p&gt;
&lt;p&gt;Euro-Q-Exa, near Munich, is the largest of the six by qubit count at 54, according to EuroHPC JU, built by IQM and coupled to SuperMUC-NG at the Leibniz Supercomputing Centre. EuroHPC JU paid EUR 10 million of its EUR 25 million cost, with Germany’s federal and Bavarian ministries covering the rest. An upgrade to 150 qubits is planned by the end of 2026, and European users reach the system through the Munich Quantum Portal.&lt;/p&gt;
&lt;p&gt;Lucy, in Bruyères-le-Châtel, is a photonic system with up to 12 qubits, according to EuroHPC JU, built by Quandela with German partner attocube. It is owned by EuroHPC JU itself, coupled to the Joliot-Curie supercomputer run by GENCI and hosting partner CEA, and cost EUR 8.5 million split equally between EuroHPC JU and France.&lt;/p&gt;
&lt;p&gt;EuroQCS-Spain, at the Barcelona Supercomputing Center, is a 10-qubit analogue neutral-atom system, according to EuroHPC JU, built by Qilimanjaro Quantum Tech and Do IT Now. It cost the same EUR 8.5 million, with EuroHPC JU funding half and Spain’s Ministry for Digital Transformation most of the rest alongside a Portuguese contribution. It is integrated with the MareNostrum 5 supercomputer for a planned three years of operation.&lt;/p&gt;
&lt;p&gt;SOL, in Bologna, according to EuroHPC JU, was the last of the original six to open, a neutral-atom system with more than 140 qubits built by Pasqal. It is coupled to the Leonardo supercomputer at CINECA and cost EUR 13 million, split equally between EuroHPC JU and Italy’s Ministry of University and Research via ICSC. A hybrid analogue/digital upgrade is planned for 2027.&lt;/p&gt;
&lt;h2 id=&quot;who-can-use-them&quot;&gt;Who can use them&lt;/h2&gt;
&lt;p&gt;EuroHPC JU opened free access to four of the six systems on 25 June 2026: Euro-Q-Exa, Lucy, PIAST-Q and VLQ. Access runs through a continuously open pilot call with monthly cut-off dates, the first on 1 August 2026. Researchers, public institutions and industry anywhere in Europe can apply at access.eurohpc-ju.europa.eu, for code development, algorithm testing and training on quantum hardware alongside classical HPC resources.&lt;/p&gt;
&lt;h2 id=&quot;what-comes-next&quot;&gt;What comes next&lt;/h2&gt;
&lt;p&gt;EuroHPC JU launched a tender on 23 July 2026 for MeluXina-Q, a semiconductor spin-qubit system that LuxProvide will host in Luxembourg. It will be coupled to the MeluXina supercomputer, start with at least 10 qubits and is planned to grow past 80. It costs EUR 11.95 million, split equally with the Luxembourg government; the tender closed on 1 September 2026.&lt;/p&gt;
&lt;p&gt;SURF, operator of the Dutch national supercomputer, signed its hosting agreement with EuroHPC JU in December 2024. Its system will be coupled to the Snellius supercomputer at Amsterdam Science Park; EuroHPC JU launched procurement for it on 15 September 2025.&lt;/p&gt;
&lt;p&gt;Separately from the EuroHPC quantum computers, the HPCQS project inaugurated two Pasqal neutral-atom simulators, Jade and Ruby, at Jülich in Germany and at CEA in France on 13 November 2025. Each controls over 100 qubits. The next milestone to watch is whether MeluXina-Q and the SURF system in the Netherlands move from procurement to inauguration.&lt;/p&gt;
</content:encoded><dc:creator>Owen Hartley</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-Quantum_computer__Germany_(17)_Z2w7A5e.jpg" medium="image"/><category>policy</category></item><item><title>AI System at Physical Superintelligence Cracks an 18-Year-Old Quantum Speedup Question</title><link>https://explorequantum.com/article/quantum-speedup-proof/</link><guid isPermaLink="true">https://explorequantum.com/article/quantum-speedup-proof/</guid><description>Physical Superintelligence says an AI research system it built produced a full proof of the 2008 Aaronson-Ambainis conjecture, a limit on quantum speedups, with Lean checking every step and Stanford researchers reviewing the result.</description><pubDate>Sat, 07 Nov 2026 10:12:33 GMT</pubDate><content:encoded>&lt;p&gt;Physical Superintelligence (PSI) says an AI research system it built produced a complete proof of the Aaronson-Ambainis conjecture, according to the company. The question about the limits of quantum speedups had stood open since 2008. The system ran on PSI’s own infrastructure under researcher David Turturean. It used instances of a non-public Anthropic model and reached the full proof on the third day of the run. Lean, a proof assistant that checks every step of a mathematical argument against formal logic, accepted it.&lt;/p&gt;
&lt;p&gt;The conjecture is about yes-or-no decision problems built from randomly chosen inputs, according to Physical Superintelligence. Its proof shows a quantum algorithm cannot gain an exponential edge over a classical one on such unstructured inputs. Doing so would require reading far more of the input data. That is a hard limit on how far quantum speedups can reach when a problem has no exploitable structure to lean on.&lt;/p&gt;
&lt;p&gt;That limit is narrower than it sounds. Physical Superintelligence says the proof bounds how many times a classical algorithm must read the input to match a quantum algorithm’s answer on most random inputs lacking a rare pattern. A quantum computer cannot save exponentially many of those reads in that setting. It does not cap the total running time of quantum algorithms generally. Nor does it rule out speedups on practical problems with exploitable structure, such as Shor’s algorithm.&lt;/p&gt;
&lt;p&gt;According to Physical Superintelligence, the system’s method was iterative rather than a single leap. It searched prior literature, proposed candidate arguments and tested them against each other. A full proof emerged on that third day. The argument was then translated into Lean and checked by its kernel. PSI had the system improve the proof’s bounds twice more, with Lean accepting each improved version.&lt;/p&gt;
&lt;p&gt;The proof now faces outside scrutiny. Physical Superintelligence says Stanford computer scientists Guy Blanc, Jordan Docter, Carmen Strassle and Li-Yang Tan have read the proof. They sent comments and are writing a shorter exposition of it for experts. PSI has published the manuscript and the Lean 4 formalisation alongside its research page. The proof has not yet been through peer review.&lt;/p&gt;
</content:encoded><dc:creator>Mara Quinlan</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-Wire_Bonded_Superconducting_Quantum_Chip_with_6_Transmons_002_Z1gCdIw.jpg" medium="image"/><category>computing</category></item><item><title>IonQ Pledges Quantum Computing Access and Satellite Data to Developing Commonwealth States</title><link>https://explorequantum.com/article/ionq-commonwealth-quantum/</link><guid isPermaLink="true">https://explorequantum.com/article/ionq-commonwealth-quantum/</guid><description>IonQ has announced three one-year commitments at CHOGM 2026 to give developing Commonwealth states access to its quantum computers, help with post-quantum migration, and satellite radar data.</description><pubDate>Sat, 07 Nov 2026 06:12:26 GMT</pubDate><content:encoded>&lt;p&gt;IonQ announced three one-year commitments at the Commonwealth Heads of Government Meeting (CHOGM) 2026 in Antigua and Barbuda, according to IonQ. They aim to give developing member states among the Commonwealth’s 56 countries earlier access to quantum computing tools, the company said on 4 November 2026. Chairman and chief executive Niccolo de Masi said the bloc’s small and developing states should not wait on the sidelines as quantum technology reshapes which economies lead.&lt;/p&gt;
&lt;p&gt;IonQ is a trapped-ion quantum computing company. Its first commitment gives universities in developing Commonwealth states an annual allocation of 300 hours of access to IonQ’s quantum computing platform, according to IonQ. That allocation also includes guest lectures from IonQ engineers and support for student hackathons.&lt;/p&gt;
&lt;p&gt;A second strand of the plan puts IonQ specialists alongside central banks and payment operators in those states, according to IonQ. Their task is to map which systems rely on cryptography that quantum computers could eventually break, and to plan the shift to quantum-safe alternatives.&lt;/p&gt;
&lt;p&gt;The third piece is a yearly allocation of synthetic aperture radar satellite imagery and analyst time, according to IonQ. IonQ says recipient states can use it for disaster response, coastline monitoring and detecting illegal fishing.&lt;/p&gt;
&lt;p&gt;IonQ describes all three commitments as running for an initial one-year period, according to IonQ. They are aimed chiefly at states in Africa, the Caribbean and the Commonwealth’s small island developing states.&lt;/p&gt;
</content:encoded><dc:creator>Owen Hartley</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-IonQ_headquarters_College_Park_MD_Z2t5Kg.jpg" medium="image"/><category>business</category></item><item><title>UK&apos;s NQCC Signs Enterprise Deal With Phasecraft to Widen Access to Quantum Software</title><link>https://explorequantum.com/article/nqcc-phasecraft-quantum-software/</link><guid isPermaLink="true">https://explorequantum.com/article/nqcc-phasecraft-quantum-software/</guid><description>The UK&apos;s National Quantum Computing Centre has signed an enterprise licensing agreement with Phasecraft, adding the company&apos;s commercial software to its SparQ Access programme for the first time.</description><pubDate>Sat, 07 Nov 2026 02:08:29 GMT</pubDate><content:encoded>&lt;p&gt;The UK’s National Quantum Computing Centre has signed an enterprise licensing agreement with Phasecraft, according to NQCC and Phasecraft. The deal brings the quantum-algorithm company’s commercial software into the NQCC’s SparQ Access programme. It is the first time the centre has added a vendor’s dedicated commercial software to that programme.&lt;/p&gt;
&lt;p&gt;Access under the deal covers three Phasecraft tools, according to NQCC and Phasecraft. Magritte handles quantum simulation for materials science and molecular modelling. Monet applies quantum-enhanced density functional theory to computational chemistry. Mondrian is an optimisation tool for problems such as logistics, grid infrastructure and finance.&lt;/p&gt;
&lt;p&gt;NQCC plans to open expressions of interest for the new access route in late November 2026, according to NQCC. The first cohort of users is due to start in spring 2027. The agreement does not disclose financial terms. NQCC has not yet said how many organisations the first cohort will admit, or how it will select them.&lt;/p&gt;
&lt;p&gt;SparQ Access is the NQCC’s programme for giving UK businesses, academic groups and public bodies access to quantum computing resources. It spares them from having to build or buy their own. Phasecraft is a quantum-algorithm company. It develops software meant to run useful chemistry and optimisation problems on today’s and near-term quantum hardware.&lt;/p&gt;
&lt;p&gt;Expressions of interest open this month. A first group of users joins the platform by spring next year.&lt;/p&gt;
</content:encoded><dc:creator>Owen Hartley</dc:creator><media:content url="https://explorequantum.com/_astro/1280px-Minister_Sir_Patrick_Vallance_officially_opens_the_National_Quantum_Computing_Centre_at_Harwell_Science_and_Innovation_Campus__Oxfordshire__United_Kingdom_on_25_October_2024_-_1_1cqoHU.jpg" medium="image"/><category>business</category></item></channel></rss>