What the Floor Told Us About Where This Industry Really Stands
I spent three days in September 2026 at Quantum World Congress, and I came home more energized than ever about this field. While it is true that no one announced a fault-tolerant quantum computer, the texture of the conversations has evolved. The pitches have changed from being mostly about qubit counts and distant roadmaps. This year, booth after booth, the relevant details were about decoders, wafers, installed systems and signed contracts.
The industry is starting to argue with itself about evidence, be it of a working machine, a real customer or a genuine advantage, and that’s exactly the kind of argument worth listening to.
I want to bring the spirit of this event back to you, conveying the genuine excitement about the moment we’re in, and pairing it with the same discipline we try to apply to every research note. It’s on all of us to separate what was demonstrated from what was merely described as a target. What follows is a walk through the three days, with a running note on which companies belong to the WisdomTree Quantum Computing Fund (WQTM) and their current weight in the portfolio.1
Three Lenses for Judging Any Claim You Hear
One framing from the event stuck with me because it’s simple enough to use every day. We can think of quantum computing progress in three layers.
- There are machines that generate better physical operations.
- There is software that turns those fragile operations into more reliable ones, including such functions as error correction, decoding and control.
- There are applications that have to prove a measurable advantage over strong classical alternatives.
Additionally, a physical qubit count, a logical state, a roadmap target and a validated customer workflow are four different things, and this year’s conference made that distinction unavoidable.
Machines Making Real, If Narrow, Progress
The most concrete news of the week came on day one. Infleqtion (5.1%) announced a live 30-logical-qubit entangled experiment across 80 atoms, a genuine encoded milestone, though it relied on post-selected loss correction rather than continuous real-time error correction, and the full methods paper is still to come.2 D-Wave (5.4%), now merged with Yale spinout Quantum Circuits, recently published a peer-reviewed result showing an entangling two-qubit gate at roughly 99.9% fidelity using “erasure-aware” dual-rail qubits, a design where the hardware itself flags where an error happened, rather than leaving software to guess.3 Riverlane, a private UK company that builds the error-decoding stack several hardware makers rely on, discussed FPGA decoder results.4 Three different companies, each with a different next scaling hurdle still in front of it.
Commercialization is happening in parallel, and it doesn’t look the same for every company. IQM (2.3%) sells and delivers installed superconducting systems today, including a first South American placement in Brazil.5 Quantum Computing Inc., or QCi (2.2%), sells room-temperature optimization hardware built around a very different physics, an open, driven photonic system rather than a universal gate-model machine. Rigetti (5.8%) is building superconducting chips out of smaller “chiplets” bonded together, and in September it both landed a finalized CHIPS Act award of up to $100 million and reported a system order from India’s C-DAC.6 None of that proves whose architecture wins; it does prove that several different business models can generate revenue before anyone reaches fault tolerance.
The Hardest Bar: Biology and Drug Discovery
The single most sobering, and, I’d argue, most exciting once you sit with it, moment of the conference was IBM’s (5.0%) update with Cleveland Clinic and RIKEN on modeling a 12,635-atom protein–ligand complex, a 2026 Gordon Bell Prize finalist.7 The team combined two 156-qubit IBM processors with a large classical supercomputing workflow, reporting more than 40 times greater molecular scale and up to 210 times better accuracy than their own prior work on a specific step. IBM was direct that the method does not yet beat the best all-classical approaches overall. That’s the right standard, and it’s a useful model for how we should read every “quantum for biology” claim that follows.
Ask whether the comparison was to a strong classical baseline, and whether the workflow was mostly classical with a quantum assist, or the reverse.
Microsoft Traveling Two Paths Simultaneously
Microsoft’s (3.3%) keynote was a study in hedging a technology bet the right way. On one hand, its Majorana 2 chip pursues topological qubits, an approach that could dramatically cut error-correction overhead if it works; the newest preprint reports parity-switching times around 20 seconds in a test structure, an interesting stability result but not yet a working logical qubit.8 On the other hand, Microsoft is already selling a hardware-agnostic logical software stack, and it has demonstrated error-correction improvements of up to roughly 800 times on Quantinuum’s (6.4%) trapped-ion hardware, real logical-qubit engineering data, running on someone else’s machine, while its own chip matures.9 That is a genuinely clever way to keep learning no matter which physical platform ultimately wins.
IonQ (6.6%), for its part, had an unusually dense September of public announcements:
- A fabricated 256-qubit Superion chip out of its newly acquired SkyWater foundry line10
- A Florida International University deployment agreement11
- A partnership with NVIDIA’s (3.1%) accelerated quantum research center12
- A widely discussed resource estimate for what it would take, architecturally, to break a Bitcoin-style elliptic-curve signature, a compiled design requiring nearly 20,000 physical qubits and 25.7 days of runtime under stated assumptions13
The Foundry and the Supply Chain behind the Headlines
Some of the most important news of the week wasn’t a qubit result at all. The U.S. Department of Commerce finalized a $1 billion CHIPS award for Anderon, an IBM-affiliated dedicated 300-millimeter quantum wafer foundry in Albany, alongside a matching roughly $1 billion IBM investment, a genuine step toward treating quantum chips as a manufacturing problem, not just a physics problem, though how much of that capacity ultimately serves customers beyond IBM is still an open question.14 It was particularly notable that the CEO of Rigetti joined the CEO of Anderon on the mainstage to talk about the importance of a specific foundry model for quantum computing.
That manufacturing story runs straight through Japan, and it was refreshing to see how much of the supply chain was represented on the floor. Fujitsu (3.7%) showed both a superconducting roadmap toward a roughly 1,000-qubit system and a newer diamond-spin photonic prototype with TU Delft.15 Hitachi (1.5%) is developing silicon-spin devices with Intel (2.5%) Japan and AIST16 under a national industrialization program. Hamamatsu Photonics (1.5%) is building the photodetectors and atom-photon interface components that several optical and neutral-atom architectures will need if they’re ever going to network smaller machines into bigger ones. None of these are headline-grabbing “we broke a record” stories, but they’re the unglamorous layer that everything else depends on.
Applications Still Have to Earn Their Advantage
QuEra, a private neutral-atom company, laid out its roadmap from today’s 256-atom Aquila platform toward a 2028 “Libra” system targeting a million reliable logical operations, built on partnerships with Amazon’s (3.2%) AWS, NVIDIA and HPE.17 Those are genuinely useful cloud and compute partnerships, and they are still roadmap commitments rather than delivered benchmarks.
The Policy Backdrop Is Real, but Read the Fine Print
Public capital showed up all over the floor, and it’s worth being precise about what each dollar actually is. Anderon’s $1 billion is a finalized award, not a proposal. By contrast, the Senate-introduced National Quantum Initiative Reauthorization Act would authorize roughly $85 million a year for the National Institute of Standards and Technology (NIST) quantum research over five years, plus smaller allocations for quantum centers and NASA R&D, a meaningful signal of political support, but an authorization sets a ceiling on future appropriations rather than committing money today.18 India’s National Quantum Mission, with an approved outlay above ₹6,000 crore (roughly $625 million USD) through 2031, is funding real infrastructure and talent programs even though its headline hardware targets remain goals rather than delivered machines.19
Conclusion: The Journey of the Quantum Computing Ecosystem Continues
The honest summary of three days is that measurable progress happened on machines, on decoders and on manufacturing capital, layer by layer, exactly as our three-lens framework predicts it should. The applications layer, including chemistry, drug discovery and optimization, remains the place where the classical bar is highest and the proof is thinnest, which is precisely why it matters most.
The range of machines, software, manufacturing capacity and applications, from small specialist firms to IBM and Microsoft, is exactly the breadth WQTM’s ecosystem-oriented approach is built to hold. The event didn’t hand us a single winner. It handed us a much clearer map of what to check next, and that’s a good trade.
1 Source: The WQTM fund page, with constituent weights as of September 25, 2026. Holdings subject to change.
2 Source: Gokhale, P. (2026, September 24). Demonstration of 30 logical qubits on Sqale. Infleqtion.
3 Source: Mehta, N., Teoh, J. D., Noh, T., Agrawal, A., Anderson, A., Birdsall, B., Brahmbhatt, A., Byrd, W., Cabrera, A., Cacioppo, M., Carroll, L., Chen, J., Chien, T.-C., Chamberlin, R. A., Curtis, J. C., Danso, D., Desigan, S. R., D’Acounto, F., Elfeky, B. H., . . . Schoelkopf, R. J. (2026). An entangling gate for dual-rail erasure qubits. Nature. Advance online publication.
4 Source: Ziad, A. B., Zalawadiya, A., Topal, C., Camps, J., Gehér, G. P., Stafford, M. P., & Turner, M. L. (2025). Local clustering decoder as a fast and adaptive hardware decoder for the surface code. Nature Communications, 16, Article 11048.
5 Source: IQM Quantum Computers. (2026, September 17). IQM to deliver its first quantum computer in South America to Brazil’s Eldorado Research Institute [Press release].
6 Sources: Rigetti Computing, Inc. (2026, September 8). Rigetti signs definitive agreement for $100M with U.S. government to accelerate R&D for superconducting quantum computing [Press release]; Rigetti Computing India P L. (2026, January 20). Rigetti announces order for a 108-qubit quantum computer from India’s Centre for Development of Advanced Computing (C-DAC) [Press release].
7 Source: Cleveland Clinic. (2026, September 9). Cleveland Clinic, RIKEN and IBM team advance to finals for 2026 ACM Gordon Bell Prize [Press release]. Cleveland Clinic Newsroom.
8 Source: Aghaee, M., et al. (2026). 20 second parity lifetime in an InAs–Pb tetron device. arXiv.
9 Source: Paetznick, A., Reichardt, B. W., da Silva, M. P., Ryan-Anderson, C., Aasen, D., Bello-Rivas, J. M., Campora, J. P., III, Chao, R., Chernoguzov, A., van Dam, W., Dreiling, J. M., Foltz, C., Frachon, F., Gaebler, J. P., Gatterman, T. M., Grans-Samuelsson, L., Gresh, D., Hayes, D., Hewitt, N., . . . Svore, K. M. (2026). Improved quantum processor logical error rates via correction and detection. Nature, 654(8118), 349–355.
10 Source: IonQ. (2026, September 8). IonQ launches Superion product line: Industry-leading, upgradeable platform designed to scale manufacturable fault tolerant quantum computing [Press release].
11 Source: IonQ. (2026, September 24). IonQ’s Superion 256 platform selected by Florida International University for flagship deployment [Press release].
12 Source: IonQ. (2026, September 23). IonQ to advance quantum supercomputing by bringing first QPU to NVIDIA Accelerated Quantum Research Center [Press release].
13 Source: Häner, T., Tripier, F., Young, J., Naehrig, M., Maksymov, A., Alam, S., Maslov, D., Parrott, M., de Sereville, Y., Sullivan, J., Webster, M., Delfosse, N., Gamble, J., & Roetteler, M. (2026). Computing 256-bit elliptic curve discrete logarithms in 26 days on a fault-tolerant trapped-ion quantum computer with 20,000 qubits. arXiv.
14 Source: Anderon. (2026, September 16). Anderon, an IBM Company, finalizes agreement with the U.S. Department of Commerce for a $1 billion CHIPS award to accelerate R&D for U.S.-based pure-play quantum foundry [Press release]. IBM Newsroom.
15 Sources: Fujitsu & RIKEN. (2025, April 22). Fujitsu and RIKEN develop world-leading 256-qubit superconducting quantum computer [Press release]. Fujitsu; Fujitsu. (2026, September 8). Fujitsu develops diamond-spin quantum computer prototype [Press release].
16 AIST refers to the National Institute of Advanced Industrial Science and Technology.
17 Source: QuEra Computing. (2026, June 15). QuEra announces 2028 fault-tolerant quantum computer and expanded multi-year strategic collaboration with AWS [Press release].
18 U.S. Senate Committee on Commerce, Science, and Transportation. (2026). National Quantum Initiative Reauthorization Act of 2026: Section-by-section summary [PDF].
19 Source: Press Information Bureau, Government of India, Cabinet. (2023, April 19). Cabinet approves National Quantum Mission to scale-up scientific & industrial R&D for quantum technologies [Press release].
