Canada's 2026 Quantum Reality Check: Milestones, Independent Referees and a First Buyer

By James Hartley | IT Insights

Start with the verdict. Physics is no longer the thing holding this sector back. Canada has spent two decades proving it can produce frontier work in quantum computing, sensing and communications. What is being tested now is far drier and far more decisive: whether that science can be converted into products with paying buyers, factories with workable tolerances, and technicians who can keep equipment running — and whether the resulting value stays in the country instead of being imported back as a finished system.

Two quieter changes explain why this particular year feels like a hinge. First, the money has changed shape, moving from open research grants toward milestone-based payouts that force a definition of "useful" into the contract itself. Second, independent referees have appeared, with published yardsticks that decide whether a machine is an instrument or an advertisement. A third change matters to readers who will never buy a quantum device: the first quantum cost most organisations will actually pay is in encryption, not computing.

Three shifts that account for the moment

1. Funding became conditional on engineering proof. The Canadian Quantum Champions Program, announced in December 2025, opened with up to C$92 million (~$64.5 million) in milestone-based contributions directed to Anyon Systems, Nord Quantique, Photonic and Xanadu, all aimed at scalable, fault-tolerant machines. Payment tied to cleared gates is a different bargain from a grant: it makes the customer's question — what can this actually do, and at what cost — a term of the deal.
2. Assessment moved to third parties. Photonic, Xanadu and Nord Quantique are taking part in an initiative led by DARPA, the Quantum Benchmarking Initiative. The program page on darpa.mil frames its own purpose in plain terms: finding out whether an industrially useful computer can be built by 2033. Stage A started with 15 companies in April 2025, and 11 were carried forward to Stage B as of 6 November 2025, per the agency's published selection. Alongside that, Innovation, Science and Economic Development Canada said in its December 2025 announcement that the National Research Council of Canada would stand up a Benchmarking Quantum Platform to run expert assessment at home rather than outsource the verdict.
3. The prize has been officially costed, twice, in two different ways. Ottawa launched its National Quantum Strategy with C$360 million over seven years committed in Budget 2021. ISED's strategy documents now describe a domestic quantum sector that could be worth roughly $139 billion with more than 200,000 jobs by 2045, while the 2025 champions-programme release projected a $17.7 billion contribution to GDP and more than 157,000 jobs over the same horizon — a wide measure and a narrow one for the same bet, and a useful reminder that forecasts in this field are arguments, not facts.

One concept unlocks the rest: the error-correction tax

Qubit counts dominate headlines and are the least informative number available. A physical qubit in a laboratory is noisy and short-lived; a logical qubit is a bundle of physical qubits organised so that errors can be spotted and corrected, and it is logical qubits that do the arithmetic. The distance between the two is a tax levied on every architecture in the field.

Nord Quantique is attacking that tax rather than out-building it. Most platforms need something like 100× to 1,000× more physical qubits than logical ones, according to chief executive and co-founder Julien Camirand Lemyre. “This is how you end up with the need for a million qubits to get to a useful quantum computer.” By folding redundancy into each individual bosonic qubit, the Sherbrooke company hopes to shrink the protective hardware wrapped around the information.

“What quantum error correction needs is not more qubits,” Camirand Lemyre said. “We need some form of redundancy.” Rival designs, he added, could cost billions to construct and sprawl to the footprint of a data center. His stated target for Nord Quantique is a machine that is genuinely useful, priced at tens of millions of dollars and fitting inside a small room. For scale, a 2025 Brookings analysis recites the commonly cited estimate that factoring a 2048-bit RSA key could demand on the order of 20 million noisy physical qubits running for roughly eight hours — a sobering yardstick for how much overhead still has to be engineered away.

Then there is the plumbing: heat, not hardware

Superconducting machines live inside dilution refrigerators, and every watt of heat leaking into the cold stage has to be pumped back out. Qubic, whose work connects computing infrastructure with possible sensing and communications uses, aims at exactly this bottleneck: catching very weak microwave signals without generating extra noise or heat. Jérôme Bourassa, chief executive and co-founder of Qubic Technologies, said the company's superconducting amplifier is designed to dissipate roughly one ten-thousandth as much heat as a conventional semiconductor amplifier. Those semiconductor units, he added, can consume as much as half of a refrigerator's cooling budget.

Release that heat and you free cooling capacity for more qubits, and the whole system becomes cheaper to own. “If we can’t make a quantum computing infrastructure that is economically sound … it won’t happen,” Bourassa said. This is the part of the story that rarely reaches a general audience: progress now looks less like a leap in qubits and more like plumbing, packaging and power budgets.

Three quantum tracks, three different clocks

Track	What it buys its owner	Who pays first	Where Canada stands	The catch
Sensing	Precision measurement of gravity, magnetic fields and time — positioning that keeps working when satellites are jammed or absent	Defence, public infrastructure and resource surveying	Julien Chosson, the Canada quantum lead at IBM, calls sensing “probably the most mature technology,”	Wins are component-sized: small contracts, long qualification cycles
Communications	Network security claims, key distribution, and sovereignty arguments about trusted links	Governments and critical-network operators	Named as a sovereign capability alongside sensing and computing in the Defense Industrial Strategy	Global revenue is often projected at $11 billion to $15 billion by 2035 (McKinsey), but standards and procurement rules lag the pitch decks
Computing	Speed-up on a narrow list of problems, mostly simulation and optimisation	Universities, cloud users and research programmes — not yet open markets	Four Canadian hardware firms are inside the national milestone programme and the DARPA assessment	“It’s still part of the research path,”

Chosson's distinction is not caution for its own sake. For him, defence can help computing by “helping build a stronger ecosystem. Until we have that, there is not much we’ll be able to develop for the defense sector on the quantum computing side.” Read plainly: the near-term value of a quantum computer to a military buyer is the supply chain and the skills it forces into existence, not the machine's output.

Why defence looks like the first realistic customer

Canada's Defense Industrial Strategy names quantum sensing, communications and computing as sovereign capabilities, which reframes the state as a customer rather than only a funder. A discussion at the Empire Club earlier this year brought military and industry leaders together around deployment and procurement, with participants pressing for companies to be connected to real operational problems instead of waiting for technology to reach full maturity. The implementation plan on the table is Quantum 2030, the Department of National Defence and Canadian Armed Forces' quantum science and technology strategy plan released in 2023, which maps a route to prototypes ready for field testing by 2030. Context matters here: Budget 2025 committed Ottawa to raising defence spending toward 2% of GDP, and the same December 2025 quantum announcement linked that trajectory to industrial policy.

For a hardware start-up, that shift is about survival, not ideology. “How do we support earlier-stage-of-development technology that sometimes has been underfunded in Canada?” asked Camirand Lemyre, who leads Nord Quantique in Sherbrooke. “I think [the Defense Industrial Strategy] is well-positioned to do just that.” Buying the box is only half the job, he stressed: “It is important for Canada to be prepared not only to buy quantum computers but also to use this technology.”

Bourassa wants a mission-driven track for sensing to sit beside the computing support. He called the Canadian Quantum Champions Program “a good start,” and told EE Times, “I don’t think it’s enough, and I think I would like to see something similar but in other areas, so think about the defense-related technologies and quantum sensing in particular.” What firms need, in his account, is capital, tighter links with the Canadian Armed Forces, and government partners willing to absorb technical failure: “Not everyone is going to win. Not everything is going to work.”

Qubic's own record shows how a purchasing arrangement can weld development to customer evaluation. In September the company announced a federal contract worth up to C$1.5 million (~$1 million) to supply nine cryogenic amplifiers with components and software for testing. That followed a C$3.5 million (~$2.5 million) seed round in June and an earlier sale of cryogenic amplifiers to Quantum Machines. Small numbers, but they are the shape a first market actually takes: a qualified test, a delivery, a repeat order.

Where the capital and the factories are going

Money is concentrating in British Columbia around Photonic. Its final investment close in May expanded a round first announced in January to more than C$275 million (~$200 million) and lifted total capital raised to more than C$475 million (~$350 million). New names in the round included BDC, EDC, Bell Ventures and InBC, joining earlier backers RBC, Telus, Microsoft and BCI — a mix of development finance, strategic telecom capital and cloud technology that says as much about risk perception as any forecast.

Photonic couples silicon spin qubits with optical links in its Entanglement First architecture, spreading computation across connected modules so that connectivity, not one monolithic chip, carries the scaling burden; its collaboration with Telus also threads quantum development into fibre already in the ground. These are routes toward commercial scale, not proof of arrival. The firm's most ambitious proposal, Project Vanguard, imagines a shared multi-tenant semiconductor facility in Canada costing up to C$500 million (~$350.5 million), serving quantum computing, AI, aerospace, defense, advanced sensing and packaging. Listed in the Canada Investment Summit Prospectus, it is designed to carry companies from prototype to production domestically, keeping intellectual property, supply-chain control and fabrication know-how at home. Related photonics work sits in the same ecosystem: among FABrIC-backed projects is Aeponyx's development of low-loss silicon nitride photonic circuits for quantum applications under Pasqal ownership, where shared fabrication and packaging turn specialist research into parts a factory can build.

The overlooked gap: preparing users before the hardware lands

Commercialisation also depends on organisations that can identify a worthwhile problem, weigh quantum methods against classical ones and wire the result into existing systems. Chosson is blunt about where the line sits: “We do not have an ROI yet. We do not have a demonstrated quantum advantage yet. We’re getting close and we have great research that’s very promising. It’s not been achieved yet.” His remedy is structural: “Right now, the best way to do this is through industry-academia collaboration. All the talent is sitting in our academia, so we really need that collaboration to spark.”

Two Canadian models are worth a general reader's attention. The University of Saskatchewan has acquired a quantum computer built on open architecture, pairing a nine-qubit Rigetti processor with control and automation technology and a dilution refrigerator made by Edmonton-based Zero Point Cryogenics — and plans to connect it to vaccine, food-security and energy expertise rather than fence it off as a physics instrument. On the access side, CMC Microsystems runs the Quantum Computing Sandbox inside the five-year, C$217 million (~$156 million) FABrIC program, bundling supported access to quantum platforms with application expertise for researchers, start-ups and small and midsized enterprises. Quandela joined in September under an agreement with CMC, adding a photonic option that includes a processor located in Canada. Sandboxes address the adoption gap that hardware alone never solves: having a machine available does not teach anyone where it might help.

Ask what the first real applications look like and the answers are unglamorous. Camirand Lemyre expects materials and molecular simulation to benefit first, citing an OLED manufacturer that wants to screen molecules before committing lab time. “It would be great if we could simulate those molecules before we actually have to spend all that time in the lab,” he said. Chosson points to Canadian work in life sciences, energy and finance, while cautioning that all of it remains exploratory.

The quantum bill that arrives early: encryption

For readers outside the industry, the most immediate consequence is a security problem with a long fuse already burning. Canada's Centre for Cyber Security guidance on the quantum threat to cryptography names the mechanism directly — “Harvest Now, Decrypt Later” — the practice of copying encrypted data today and holding it until a sufficiently powerful machine exists to open it. That means secret-but-delayed exposure, not tomorrow's breakage, and it puts hospitals, banks, law firms and government archives under pressure to inventory their encryption now. Canada's financial regulator, OSFI, has published staged quantum-readiness timelines asking institutions to assess this exposure and evaluate migration paths; commercial roadmaps converge on the same urgency window, with Cloudflare publicly targeting 2029 for full post-quantum protection. Labour markets register the same shift from a different angle: MIT's Quantum Index Report 2025 finds the share of U.S. job postings requiring quantum skills has almost tripled since 2018, evidence that the scarce asset is people who can bridge physics and operations, not machines.

Five misconceptions that distort the debate

• “A quantum computer is a faster computer.” It is a different instrument for a narrow class of problems; everyday workloads will run on classical hardware, increasingly with quantum devices called through the cloud.
• “More qubits means more power.” Without error correction, extra physical qubits add noise. The meaningful number is logical qubits, and the meaningful ratio is value delivered versus cost to deliver it.
• “Defence money means the technology is ready.” Procurement is being used to pull earlier-stage work forward and to build a supply chain, which is the opposite of a maturity signal.
• “Funding announcements equal commercialisation.” Milestone programmes pay on cleared gates; a headline round says nothing about factory yield, service contracts or replacement parts.
• “There is nothing to do until machines are useful.” Encryption migration, data-retention review and application discovery are all decisions with present-day deadlines.

Assembling one national industry instead of several showcases

The semiconductor overlap is where the argument gets concrete. Rafal Janik, chief operating officer of Xanadu in Toronto, notes that quantum investment builds chip capability as a by-product — cleanrooms, tooling, packaging processes and trained staff — value that outlives any single architecture, provided institutions stop duplicating one another's work. Coordination need not mean closing ranks: at the Stewart Blusson Quantum Matter Institute at UBC, quantum-materials research is wired to partners in Germany, Japan and elsewhere.

Camirand Lemyre rejects the single-hub idea outright. Sherbrooke, Waterloo, Toronto and British Columbia bring different strengths, he argues, and Canada needs both an early-stage pipeline and the production and engineering workforce for the mature phase. “Whoever builds a quantum computer that is useful is a global company. … It’s not about one single hub; it’s really about building a national champion.” For Bourassa the unfinished business is capital that keeps decades of Canadian research from being monetised elsewhere: “Now it’s a matter of how we make sure that we can create the best possible value from all the research investment that has been done over the past several decades in Canada and making sure that it is retained in Canada,” he said. Government support, in his view, is good and instrumental but can strengthen considerably as competition rises. Camirand Lemyre named the cultural blocker: “We have not been a country that is known to take large risk on their companies.”

What different readers should actually do

1. IT and security leads: inventory where long-lived secrets are encrypted and how; ask every vendor for a post-quantum cryptography roadmap and evidence of crypto-agility; treat key exchange, not bulk storage, as the first migration target.
2. Business and procurement teams: run problem-first discovery. Use the Quantum Computing Sandbox and university partnerships to test one economically relevant use case instead of buying a machine to justify.
3. Founders and component suppliers: register on federal supplier lists early, target small qualified tests in the Qubic pattern, and read milestone terms as product requirements.
4. Investors and policy readers: watch three dates — 2029 for the cryptography migration crunch, 2030 for field-testable prototypes under Quantum 2030, and 2033 for the benchmark test of whether computational value beats cost.
5. General readers: when a quantum headline cites qubit counts, ask instead about error correction overhead, heat load, packaging and who pays.

Frequently asked questions

Is Canadian quantum research world-class? By most scientific measures, yes — which is precisely why the policy argument has moved on to manufacturing, procurement and workforce rather than discovery.

Will quantum computers break encryption next year? Almost certainly not, but data copied today may be readable later. That “Harvest Now, Decrypt Later” risk, as Canada's cyber-security guidance names it, is the reason migration planning starts now.

Why is defence involved if computing isn't commercial yet? Because sensing and communications deliver value sooner, and because a patient institutional buyer builds the supplier base, test facilities and engineering talent that computing will need.

What are the first useful applications likely to be? Materials and molecular simulation, including screening candidates before lab testing, with life sciences, energy and finance as exploratory fronts.

How would we know if Canada succeeded? A machine sold and serviced domestically at tens of millions of dollars, a shared fab taking prototypes into production, milestone programmes paying out rather than clawing back, and a measurable share of the value staying in the country.

The transition is neither a foregone conclusion nor a fairy tale. Canada has the science, an unusual set of referees willing to publish hard criteria, a state customer learning to buy earlier than it once did, and capital that is finally willing to fund factories as well as laboratories. What remains unproven is the middle of the curve: the error-correction overhead, the heat budget, the packaging plant and the trained workforce that decide whether a national research strength becomes a national industry. The next test is whether funding programmes, procurement openings and shared facilities can carry that crossing — not merely producing discovery, but keeping the capabilities and the economic value it creates at home.