Xanadu Quantum Technologies Limited Class B Subordinate Voting Shares Q1 2026 Earnings Call Transcript

Quartr
Provided by Quartr
May 14, 2026

Key Takeaways

  • Positive Sentiment: Xanadu completed its public listing on Nasdaq and the Toronto Stock Exchange, generating $302 million in gross proceeds and improving its funding runway for the next phase of development.
  • Positive Sentiment: Management said the company now has CAD 272 million in cash and is pursuing additional Canadian government funding and a CAD 300 million synthetic ATM facility to support growth with what it called a relatively less dilutive financing path.
  • Positive Sentiment: The company highlighted continued technical progress in photonic quantum computing, including a 60% year-over-year reduction in optical loss and Aurora’s status as a modular, scalable, networked system with 12 logical qubits.
  • Neutral Sentiment: PennyLane remains a key strategic asset, with management citing 35,000 active users and 200,000 monthly downloads, plus partnerships with universities and enterprises that could help drive future monetization.
  • Neutral Sentiment: Near-term financials remain loss-making as the company ramps spending: Q1 revenue rose to CAD 2.8 million, but adjusted EBITDA loss widened to CAD 13.9 million and management expects expenses to increase further as engineering and wafer activity accelerates.
AI Generated. May Contain Errors.
Earnings Conference Call
Xanadu Quantum Technologies Limited Class B Subordinate Voting Shares Q1 2026
00:00 / 00:00

There are 11 speakers on the call.

Speaker 6

As a reminder, this conference is being recorded for replay purposes. I would now like to turn the call over to Brett Harris, Vice President of Investor Relations. Please proceed.

Operator

Thank you, and good afternoon, everyone. Welcome to Xanadu's inaugural quarterly earnings conference call. Joining me on the call today are Dr. Christian Weedbrook, Founder and Chief Executive Officer, as well as Michael Trzupek, Chief Financial Officer. Further, this call is being webcast live and will be made available for a period of time on Xanadu's website. This call contains time-sensitive information that is accurate only as of the date of this live webcast of this call, May 14, 2026. During this call, we may make forward-looking statements based on current expectations. These statements are based off management's current beliefs and expectations as of today and are subject to a number of significant risks and uncertainties and our actual results may differ materially.

Operator

These risks and uncertainties are discussed in the company's periodic reports filed with the SEC and should be referenced by anyone considering making any investments in the company's securities. Xanadu disclaims any obligation to update any of these statements. Now, I'd like to turn the call over to Dr. Christian Weedbrook, Founder and CEO of Xanadu. Christian.

Speaker 1

Thank you, Brett. Good afternoon, everyone. I'm Christian Weedbrook, founder and CEO of Xanadu. I started this company in 2016 with a white paper and a conviction that photonics was the right path to building a scalable quantum computer. After a decade of research, peer-reviewed breakthroughs, and now a public listing on Nasdaq and the Toronto Stock Exchange, I'm more convinced of that than ever. For those newer to our story, Xanadu is a Canadian quantum computing company with a single driving mission to build quantum computers that are useful and available to people everywhere. We are a full-stack quantum computing company, meaning we develop both the photonic hardware and the software needed to run it.

Speaker 1

We believe our approach, which uses photons or light particles, gives us a distinct and durable path to building a large-scale quantum computer and a quantum data center that can solve some of the world's most complex and pressing challenges. I want to take a moment to acknowledge what this call represents, not just a financial reporting milestone, but a public commitment to transparency, to accountability, and to delivering on our technological roadmap for building scalable, fault-tolerant quantum computers. I'll spend a few minutes on Q1 highlights and the state of the business before handing it over to Michael for a deeper look at the financials. On March 27, 2026, Xanadu became a publicly listed company on both Nasdaq and the Toronto Stock Exchange under the ticker symbol XNDU, following the completion of our business combination agreement with Crane Harbor Acquisition Corp.

Speaker 1

Our public listing was more than a capital event. It was a signal to the scientific investment communities that photonic quantum computing has moved beyond the experimental R&D stage into one of scaling, where progress depends less on theoretical physics and more on practical engineering. The go-public transaction generated $302 million in gross proceeds, which, combined with approximately CAD 285 million of anticipated Canadian and Ontario government funding currently under negotiation, accelerates Xanadu's path to building a quantum data center by 2029, 2030. We hosted our inaugural Analyst Day on March 4 in New York City, where we showcased technical progress, roadmap updates, and a deeper view of our commercialization strategy. For those who joined us, thank you. For those who weren't able to attend, the materials are available on our investor relations website.

Speaker 1

Before going deeper into our first quarter highlights, I'd like to give a broader overview of our technology for those new to Xanadu, its differentiators, and our progress to date. Every decision we make is made with one goal in mind, to build quantum computers that are useful and available to people everywhere. Let me step back and frame our strategy. At Xanadu, we believe photonics is one of the most credible path for scaling quantum computing. We will see five main reasons why. First, our system architecture is modular and networked, allowing us to scale systems through interconnects, much like modern data centers do today. Second, we leverage existing semiconductor and photonic supply chains rather than building a bespoke manufacturing ecosystem from scratch. Third is room temperature computation. Every approach needs some form of cooling, whether it's cryogenics or laser cooling.

Speaker 1

At Xanadu, we need cooling only to initialize or turn on our quantum computer, and after that, the qubits, gates, and measurements are all performed at room temperature. Fourth, photonics enables real-time and flexible error correction. This gives Xanadu greater freedom to implement advanced error correction schemes that are difficult to realize in other architectures. In particular, our error correction supports low overhead approaches to reduce the resources required for fault-tolerant quantum computing. Finally, photonics offers a fundamental advantage in gate speed, with operations occurring at frequencies of 100 megahertz. This is anywhere from 10 to 1,000 times faster than other modalities. Ultimately, two primary barriers stand between the industry and a quantum data center, improving the performance of quantum computing modules to reach fault tolerance and connecting those modules through scalable network systems. The good news is we've already put the networking challenge behind us.

Speaker 1

Our Aurora system, as demonstrated in Nature, is the world's first modular, scalable, and networked photonic quantum computer, featuring 12 logical qubits across 35 photonic chips and 13 km of fiber optics, computing at room temperature with real-time error correction decoding. With the networking problem solved, our central focus is on improving the performance of our photonic chips by reducing loss through engineering improvements. We continue to make real progress. In 2025, we reduced optical loss by 60% year-over-year. This was a step function improvement, and we're continuing to drive progress. We are targeting a qubit factory by 2026-2027, fault-tolerant operations by 2028, a quantum data center in 2029-2030, and scaling toward up to 500 logical qubits between 2029-2030, while advancing our manufacturing capabilities to match. We don't stop at the hardware level.

Speaker 1

As we've seen in companies like Nvidia, hardware alone doesn't build transformational companies. It's a software ecosystem on top of that that unlocks the full utility of hardware and keeps customers coming back. That's why we developed PennyLane, an open source quantum software platform to be the second major plank of our company strategy and a long-term competitive advantage. PennyLane is how developers, researchers, and enterprises access and explore quantum computing today and prepare for the fault-tolerant hardware of tomorrow. So much so that we have 150 university partners across the world, including University of Toronto, Johns Hopkins University, and the University of Maryland. Our 3-year partnership through UMD's National Quantum Lab is focused on workforce development and commercialization across the Mid-Atlantic. We are training the next generation of quantum scientists and engineers to program our platform.

Speaker 1

These developers who learn quantum computing on PennyLane today become the customers, partners, and colleagues of tomorrow. PennyLane is one of the most widely used quantum software platforms. The adoption growth we are seeing over 35,000 active users and 200,000 monthly downloads as of the Analyst Day tells us that the quantum developer community is expanding rapidly and Xanadu is at the center of it. While it's free today, enterprise engagements like our research initiative with Lockheed Martin to advance the foundations of quantum machine learning opens a path for monetization down the road. On the Canadian side, in March, we announced negotiations for up to CAD 390 million from the governments of Canada and Ontario to launch Project OPTIMISM and establish advanced semiconductor and photonic manufacturing infrastructure supporting Canada's quantum supply chain.

Speaker 1

These discussions are ongoing, and we look forward to updating investors as they progress. In the quarter, we announced significant new partnerships with AMD, Lockheed Martin, TELUS, and Fidelity Center for Applied Technology, advancing commercialization across defense, finance, and telecommunication. These build on our 2025 partnerships with Mitsubishi Chemical, Rolls-Royce, Riverlane, Corning, and Applied Materials. This ecosystem is intentional. Quantum will not succeed in isolation. Close collaboration with end users is how we turn algorithms into real-world applications and ensure we are building practical, commercially useful systems. For example, with AMD, we're advancing hybrid quantum classical computing for aerospace and engineering simulations using GPUs to run 20 qubit, 35 million gate quantum computational fluid dynamics simulations and accelerating workflows by 25 times versus CPUs.

Speaker 1

While our client and research partnerships are critical to driving useful applications and demand, our large-scale hardware manufacturing partnerships such as AMD, Corning, and EV Group are a key structural advantage. These partners bring deep hardware engineering expertise and enable rapid iteration at scale by leveraging proven high-volume manufacturing processes and existing production lines. This allows us to accelerate development cycles, reduce execution risk, and deploy capital more efficiently, ultimately enabling us to invest more into chip production while lowering cost per iteration. This combination of photonics-based hardware innovation and a user-friendly software platform creates a foundation of our strategy to drive near-term utility while maintaining a credible path towards fault-tolerant quantum advantage. Our mission is to build quantum computers that are useful and available to people everywhere, and that mission guides every investment we make. Now, let me pass it on to Michael to discuss our key financial highlights.

Speaker 4

Thanks, Christian. Before I walk through our quarterly performance, I want to highlight that I joined the company to help scale a disciplined financial and operating framework that supports our long-term roadmap through 2030 and beyond. While our first priority over the next few years is on progress against the roadmap Christian laid out and achieving up to 500 logical qubits in 2029 and 2030, we are also laying the foundation for revenue-generating opportunities through a multi-pronged structure, which includes subscription-based cloud access or quantum computing as a service, QCAST, a SaaS type revenue model for enterprise-grade PennyLane, on-premise system sales, IP licensing of our photonic subsystems, and the potential monetization of strategic partnerships. Turning to our financial position, we ended the quarter with CAD 272 million of cash.

Speaker 4

Our go public transaction generated $302 million in gross proceeds, which combined with approximately CAD 285 million of anticipated Canadian government funding, meaningfully accelerates our path toward building a quantum data center by 2029, 2030. This capital base provides us with the flexibility to execute against both our technical and commercial priorities. It is also important to note that the financing support from the Canadian government will not appear as cash on our balance sheet up front. Instead, those funds are expected to be received over time as qualifying investments as R&D activities occur. We are planning to put a CAD 300 million synthetic ATM facility in place, which would give us the flexibility to raise capital over time as opportunities present themselves.

Speaker 4

After evaluating a range of financing alternatives, we believe this structure provides what we expect to be the most cost-effective and least dilutive path to funding the business over the long term. To be clear on this, the synthetic ATM shares will be a primary issuance and all proceeds will flow directly to the Xanadu balance sheet, increasing our cash reserves. Our approach will be disciplined and opportunistic, any capital raised would be used to support the continued development and scaling of our quantum computing platform and technology roadmap. Importantly, access to capital allows us to innovate faster. We are accelerating commercial chip runs with our foundry partners, increasing iteration speed and improving device performance on a shorter cycle.

Speaker 4

In parallel, we are scaling our engineering organization with a deliberate shift towards hiring more systems, software, and hardware engineers alongside our world-class physics team to support the transition from research to product implementation. Let's move to our financials. Revenue increased to CAD 2.8 million from CAD 700,000 in the prior year quarter, primarily driven by revenue recognized from our DARPA QBI Stage B participation. We also recognized approximately CAD 5 million of grant income related to the Canadian Quantum Champions Program and other operating income. We believe the support from both DARPA and the Canadian government is a strong validation of our research, technology roadmap, and the commercial and operational potential of our platform.

Speaker 4

R&D expense increased to approximately CAD 17 million from CAD 10 million in the prior year quarter, reflecting continued investment in engineering talent, increased wafer iterations made in partnership with our foundry partners, and expanded process development activities. G&A expense was approximately CAD 10 million during the quarter, including roughly CAD 5.5 million of non-recurring costs associated with our recent SPAC transaction and public listing on the Nasdaq and Toronto Stock Exchange. Adjusted EBITDA loss was CAD 13.9 million compared to a loss of CAD 10.6 million in the prior year's quarter. We expect spending to increase in subsequent quarters as we continue to add engineering talent and accelerate our investment in R&D through wafer and chip production. I want to quickly clarify our share count as of March 31st, 2026. We ended the quarter with 298 million shares outstanding.

Speaker 4

255 million are Class A shares and 43 million Class B shares. Of the issued shares, over 250 million are held by legacy Xanadu shareholders and remain subject to the 180-day lock-up provisions established in connection with our SPAC transaction. These shares will not be freely tradable until September of this year. With a strong capital position, growing support from both government and commercial partners, and a disciplined approach to scaling the business, we believe we are well-positioned to execute against our technical roadmap and long-term commercialization strategy. Our focus remains on accelerating innovation, expanding our engineering capabilities, and advancing towards large-scale fault-tolerant quantum computing while thoughtfully managing capital deployment and maintaining financial flexibility. We're excited about the progress we're making and the opportunities ahead. With that, operator, we'd now like to open the call up for questions.

Speaker 6

Thank you. We will now begin the question-and-answer session. The first question comes from Nehal Chokshi with Northland Capital Markets. Please go ahead.

Speaker 5

Thank you, and congrats on your first earnings call. It's exciting. PennyLane, the Beatles song, runs through me here. 35,000 monthly users, how does that compare to a year ago? Can you clarify who do you view as PennyLane's closest competitor? Where do they stand in terms of monthly users as well?

Speaker 1

Yeah. Thanks, Nehal. We can get back to you on the specific how does it compare. We can say broadly, though, that it has been increasing since, you know, averaged over a seven, eight-year period for sure. That ties in nicely, I guess, with your second part. The two top ones, according to, you know, surveys and just working with partners and folks is really PennyLane and Qiskit, that's IBM's software offering. What we do like about PennyLane, though, is it actually is if not most widely used agnostic platforms anywhere in the world. It works on, as you know, many different hardware providers, not just the photonic-based approach, but, you know, they're the top two ones.

Speaker 4

I would also throw in Nvidia, they're late to the game, but doing some great work as well. PennyLane and Qiskit are for sure up there.

Speaker 5

Great. Thank you. Just to be clear, is Qiskit a hardware agnostic platform as well?

Speaker 1

That's a good question. I would phrase that as it's predominantly on superconducting qubits. It's IBM's one, as you know, and it's really about promoting and getting as many people using superconducting qubits. There are some other platforms where it is offered on. To our knowledge, nowhere as widely as agnostic as what PennyLane is.

Speaker 5

Got it. Okay. On the hardware side, can you give us any visibility into the linearity on the reduction of the optical loss from, I think, it was now 23 DB to, you know, target around 1 DB or so over, I think, a 3 to 4 year time frame here?

Speaker 1

Yeah, it looks linear, but it is something that over time you often work very, very hard to get. Where we are now is incremental improvements. You can see since I think that it was since 2021, 2022, up to 200 times reduction in one path. You're not gonna see 200 times reduction from where we are now 'cause we're getting very close to this line, which is normalized to one. That's when you hit threshold, and you need to go a little bit below that to be, as the name suggests, below threshold. That's where you can start solving important customer problems. One of the big reasons why we raised and went public is the fact that we can do more and more chip runs.

Speaker 1

These chip runs are really one of the keys to reducing loss. Often, you know, it's kind of quasi-linear if you extract it out. It looks that way, you know, chip runs can take anywhere from 2 to 5 months. Once we get those back, we test and measure the wafers. We understand what improvements we need to make beyond that, we do update our chip design, send them back off to different foundries. A lot of it is through chip runs. Other aspects where we'll see improvements and have seen in the past are fiber to chip coupling as well. Reducing those losses, you know, how you input and extract light off the chips are very important.

Speaker 1

We're very confident to achieve for us getting below that line, hitting that line and going below it. It only makes it more certain, having raised quite a bit of money by going public.

Speaker 5

All right, great. I'm gonna pass it along.

Speaker 1

Thank you.

Speaker 6

The next question comes from Kingsley Crane with Canaccord. Please go ahead.

Speaker 3

Hi. Thanks for taking the question. Would echo my congrats on a successful first earnings call. On the NVIDIA Ising announcement, certainly validates the quantum sector. It helped lift the quantum sector. You know, I think the underlying tools are more modality agnostic, but photonics more of an architectural advantage and clock speed, different error correction mechanism than, you know, superconducting or neutral atoms. Would just love to get your thoughts on that announcement in general, and then just like from a fundamental standpoint, how you think photonics could or could not benefit from what the work NVIDIA is doing there.

Speaker 1

Yeah, that's a really great question. You know, one of the things that we've said we'll do to the best of our ability from day one, and this kind of answer will give you the, that flavor, is really give you an idea of what's real and what's not out there. It's great to see Nvidia in this space. They've been working hard for a couple of years now on a variety of things. Ultimately, their goal, as they've told us, is to sell more GPUs, which makes complete sense. Companies in the quantum space wanna use the GPUs for simulations of algorithms. For benchmarking. They also wanna use it for error correction and fault tolerance.

Speaker 1

Now there seem to be the stocks including Xanadu, whence people kinda look for patterns and say that a lot of the stocks went up because of that Nvidia announcement. You know, we don't wanna lift, look a gift horse in the mouth, it was great to see that. The substance behind it, I don't think, to be fully honest, really, warranted those stocks going up the way they did. I think it was more the fact that Nvidia actually is doing more in this field is a great part. The actual what they offered from a Xanadu point of view, you know, wasn't as interesting for us even though it was touted as being agnostic.

Speaker 1

The reason is, it's a good reason, is those tools they released, we've had them internally now for 5 or 6 years for our photonic system. That was one aspect. The other aspect, which is a really, really important point here. A lot of the GPUs, when you build a large scale quantum computer, you're gonna have a lot of quantum stuff, but you're also gonna have a lot of classical compute. The other approaches compared to our photonic and Xanadu's photonic-based approach are significantly slower. Because they're slower, they can actually use NVIDIA's GPUs for the error correction part. We cannot use NVIDIA's GPUs for our error correction because we're too fast, which is a really amazing thing. Our clock rates are faster than definitely ion traps for neutral atoms, up to, I think, 10,000 times faster.

Speaker 1

To look at that, think of a problem that those guys can solve. They solve it in a year, fantastic, 'cause it would've taken millions of years. Then we can solve the same problem in roughly eight hours. I think over time, that's gonna be our real killer, one of our killer, you know, benefits of using photonics. What that means is the GPUs are too slow because we're very fast, so we have to resort to FPGAs. You would've seen, I think just before we went public, one of the great announcements over the last few months for us with AMD, and so working with them because FPGAs are far faster for this particular task. That's why, you know, that's how what we do relates to NVIDIA's announcement.

Speaker 3

Thanks. That's helpful to dispel some of those misconceptions. On the financial side, this is, you know, both for Christian and Michael, but, you know, you've talked about not chasing non-recurring engineering revenue and services that could divert talent. You've been pretty successful with the professional services the past couple quarters. Just would be helpful for investors, like where is that line as enterprise interest builds? Like how do you balance accepting versus declining those kinds of engagements? Just a reminder on linearity, what we could expect for that line the rest of the year. Thanks.

Speaker 1

That's a really good question, and Michael and I kind of phrase it, we're after partnerships, not customers for the most part. Now, some of those partnerships will inevitably and do become customers. What we do, we think long-term here at Xanadu. The long-term view is 2029, 2030. Let's picture we're there already. We have our large scale quantum computer that we're calling a quantum data center, has hundreds up to 500 logical qubits doing everything that you'd want this computer to do. You picture that, it's access over the cloud, and we'll have, you know, a subset, a small key group of customers at that point that we're using it, you know, drug discovery, pharmaceuticals and so forth.

Speaker 1

You work backwards and say, "Well, how do we get to that point?" We get to that point by being focused on hardware milestones. We get to that point by being selective in our partnerships. You know, we've announced a few partnerships over the last few months. Mentioned AMD. Lockheed Martin is a really great one as well. These are partners that hopefully will take with us over the next few years until we hit 2029, 2030, and potentially be some of the first customers of this. We really look at partners rather than customers. Having said that, Lockheed Martin is a paid partnership, so we do get the best of both worlds there.

Speaker 1

I really think that by us thinking long-term in terms of revenue, the eye on the prize is key, that the pie is far orders of magnitude bigger when you have a large scale quantum computer rather than selling, you know, a quantum computer for a few million CAD, where GPUs can really simulate the same answer. Ultimately takes, as you mentioned, takes our best talent off the prize. That's kinda how we see it.

Speaker 6

The next question comes from Todd Coupland with CIBC. Please go ahead.

Speaker 8

Great and good evening, everyone. Christian, I wanted to ask about DARPA Stage B. Could you just give us an update on how that's going and when we might expect the decisions for Stage C?

Speaker 1

Thanks for the question, Todd. No real new updates. As you know, this is a year-long Phase B. It could be sooner, it could be later, depending on what DARPA wants and how any of us deliver that. What I can say is we'll know more towards the end of the year. Our year is up around November. As mentioned, it could be earlier or it could be later. They have, as other companies have come on site on Xanadu, and so far it's, we have a great rapport and collaboration with DARPA. That's as much as what we can say, and honestly, as much as what we know. The team still has a few more months worth of work and deliverables.

Speaker 8

Yeah. When that outcome happens, how significant of a validation is it?

Speaker 1

That's a really good question because Rigetti's doing some great stuff, and from my knowledge, they didn't get in or didn't get in straight away. Doesn't mean they won't get in later on. I don't know how much of effect that had on them, so that was kinda one data point. I would say, though, that it's hard to tell. You kinda think maybe you may be punished by not getting in or maybe you're rewarded by getting in, but it's so early to tell. I would say that it's great that we're in and we made it to Phase B, and Phase C, the carrot has always been CAD 300 million. I honestly don't know, and I don't think many people know what the answer will be.

Speaker 1

It's really hard to tell the market for quantum companies, how things are reacting. I would say, you know, based on $300 million, it would be great to get in, and it's not like we don't wanna get in, and we're doing our best to get in. We'll see. It's probably a quick, nice segue to sort of remind everyone, perhaps on the call that we are in final negotiations of, you know, independent of that $300 million, final negotiations of CAD 390 million. That's looking good. No extra updates there. Still final negotiations. And then also we do have the Canadian QBI, called the Canadian Quantum Champions Program, which the government has said they wanna match what's going on in the $300 million plus QBI program.

Speaker 1

To our knowledge, best knowledge, that's still going ahead. We have many opportunities, independent of what happens, in how we or others proceed, to Phase C.

Speaker 8

If I could just ask on the OpEx, I guess the run rate ex the non-recurring items is what? Around CAD 21 million. What should that grow at, as we go through 2026? Thanks very much.

Speaker 4

Yeah. As we've talked about, we intend later this summer to provide guidance on this and other engineering metrics. We're not prepared to do that today. I'll go back to what I've shared with you in the past, which is obviously as we raised this money, it was to accelerate our engineering growth and our wafer starts. You can anticipate that it's going to grow over the year, and we'll be able to provide more specifics later this summer.

Speaker 8

Great. Thanks, Michael.

Speaker 6

The next question comes from John McPeake with Rosenblatt Securities. Please go ahead.

Speaker 2

Thank you. Nice work on going public, Christian and Michael and team, and welcome to these quarterly calls. Can you remind us how many silicon turns it'll take to get to your targeted loss rate and how long those turns take? Will the EV Group partnership help speed that up in any way? I saw that cross the other day.

Speaker 1

Thanks, John. Probably not, in terms of the last question, last part of the question. The first part of the question, to be fully honest, it's hard to know exactly how many chip runs, and there's a couple of reasons why. It's a very complex process. What we do is we do all the photonic integrated chip designs ourselves here in Toronto, send them off to different foundries around the world. That's where it gets a little bit tricky because we have three or four different substrates that make up our computer, and what that means, silicon nitride, lithium niobate and others. That means that we work with the best foundries that can do each of these particular materials.

Speaker 1

Not only do we have multiple chip runs because of the different materials, we have multiple foundries for each of the materials as well. To add to that, we actually have different levels of production. We work, for instance, we work with a foundry, university foundry in Waterloo. That's just for prototyping, which allows us to understand our chip designs much faster to sort of feed into the large foundries. We work with medium style foundries, dedicated foundries like say, NY CREATES in Albany. We also work with large foundries like Tower and so forth, and have worked with GlobalFoundries as well. UMC is another one for large scale manufacturing. Applied Materials is another exciting one we're working with. A lot of them are applying to 300 mil tools as well.

Speaker 1

When you look at the tools, we're doing some on 200 mil, 300 mil. Overall, I would say that's probably a better way to understand it, rather than how many runs. The more that we do and the faster we can do it, we'll get to the loss reduction faster. Anyway, that's how we think about it.

Speaker 2

Okay. The qubit factory, could you just elaborate a little bit on what that means in the roadmap?

Speaker 1

Yeah, for sure. You know, I always come back to its name is good because it indicates, you know, some sort of factory, meaning that we're making a lot of something. The thing that we're making a lot of are qubits. Now, our qubits are photonic qubits. More specifically, they're called GKP qubits. It's just a name named after the authors who came up with the original paper. Essentially, it's a type of photonic qubit. We demonstrated one, probably now coming up to getting close to, you know, maybe 10, 11 months ago, it was published in Nature. Very great for our approach because it allows you to tolerate losses, which is this thing that we're trying to reduce. If we find ways of tolerating them at certain levels, that's just as good as well.

Speaker 1

They're the type of qubits, and this factory is a way to create as many as that we want. Once you have them kinda on demand, you can now use them to sorta do computing. The reason why we broke it up that way, another reason, not only because of its importance, but it, you know, it's the toughest part, the final tough part of our computer, meaning that once we solve that, and we're well on the way to solving that, everything else is an order of 2 magnitude easier to achieve. A lot of the parts that we had to master for the qubit factory. Think of that as kind of a nucleus or core that once we solve that, everything else kinda comes out of those results. Yeah, that's essentially what it is.

Speaker 1

Beyond that, from perhaps what you're looking at is the fault tolerance. We have these qubits. We have a factory making them now, we need to make them even better, meaning loss reduction. Really the theme of loss reduction transcends through all the years. As mentioned with our Aurora computer from last year, everything is there to really change the world. The computer structure is there. You even have a qubit factory, kind of an MVP version. That's why we need to kind of keep developing that. It really is a loss reduction exercise.

Speaker 1

You need to obviously put all the parts together, even if you achieve the loss reduction, say today, for instance, you still need to build a computer and that you have simultaneously component improvements, but also assembly happening over the years as well.

Speaker 2

Yeah, that's helpful. Thank you. Good luck, guys.

Speaker 1

Thank you.

Speaker 6

The next question comes from Quinn Bolton with Needham & Company. Please go ahead.

Speaker 7

Thank you for taking my question. Christian and Michael, congrats on the going public transaction. Christian, maybe just start with you. Just wanted to ask, you mentioned, you know, increasing costs for the chip design and getting more wafers through the foundries. How much of the improvement to reduce loss comes from chip design versus process node or process technology improvements? I think at Analyst Day, you mentioned things like you need to improve surface roughness, which sounds like it's, you know, etch and depth, you know, kinda related rather than just how do you design the component. Maybe how much of it, you know, comes on, you know, design versus process technology?

Speaker 1

Yeah. As you alluded to, it is a bit of both. It's hard to kinda give a percentage. I would say it weighs more heavily towards the chip designs themselves. The reason there is that we're using photonics, and we can really stand on the shoulder of the telecommunication industry, and that's why we can work with UMC, Tower, and others because they've been working with photonics for probably decades now. What that means is we can use a lot of their tools already and the nodes are already well-established.

Speaker 1

Having said that, we do have what we call our fabrication process engineering team that work on the ground with these foundries that can offer suggestions on how to, you know, change the material structure a little bit, but nothing where we have to move to a whole new material, for instance. We use lithium niobate that's well-known and well-used for transceiver market, Datacom and so forth. In order to reduce the loss more, we have to do some tweaks on the process development, but it's not like we have to buy, at least not at the moment, a brand new tool or change it from lithium niobate to something else. That's really key. The other thing that's really important we're seeing too, and, you know, Applied Materials is a good example.

Speaker 1

The latest tools and processes are often characterized by, you know, saying that you're using 300 mil tools. More and more of the stuff we're doing is on 300 mil rather than 200 mil. We still do some on 200 mil, but the good news there is that you're using the most advanced tools that already exist. We didn't have to buy, you know, create them. Those advanced tools will help us to reduce the loss as well because they're state-of-the-art, essentially.

Speaker 7

Got it. I wanted to follow up on PennyLane. I think you'd mentioned the monetization plan for that down the road would be sort of enterprise licenses. Wanted to know, does PennyLane allow sort of compilation all the way down to a specific hardware platform? If it does, is there an opportunity at any point to do more of a usage-based model for PennyLane, where you can sort of, you know, track the usage of hardware resources and charge accordingly?

Speaker 1

Stay tuned on that last part. Hopefully, we may have a good announcement on that soon. That relates to the general idea that we didn't give too many details, but just, you know. As anyone that's had open source, it's a common thing to think about how to commercialize this. We're looking into that. Stay, you know, please stay tuned on that. You know, getting back to your first point, the software stack at the very top is really, really agnostic, like PennyLane. It's like you type in a command, like, I don't know, you wanna do Shor's algorithm, whatever it is. You don't really care per se about the hardware. As you travel down the stack and at the very opposite end, you need software that controls the hardware.

Speaker 1

It's a gradient. As you go from top to bottom and ultimately at the bottom, it really has to suit the specific hardware. We have internal software and tools that is really related to our photonic chips. That also means our photonic chips. There are other photonic companies doing great work, but it wouldn't even really be adaptable to theirs. It's very specific. That's true, I think, unless you're all doing superconducting qubits or all doing ion traps, you really got to stick into, you know, in your modality. Having said that, as you go down from PennyLane, you do cross into the compilation part, which you mentioned. As part of that, a subset, we have Catalyst.

Speaker 1

Catalyst is, to our best knowledge, one of the most widely, if not the most widely used compilation software out there. It is agnostic, so meaning we can use it, plus others can use it on their hardware as well. I think that's when you go below that even more, then you start heading to the territory where it has to get even more specific. That's how we think about it. We've been promoting Catalyst a lot, and we're looking to do more and more of that, along with kind of the larger package, which is PennyLane.

Speaker 7

Excellent. Thank you.

Speaker 1

No problem. Thanks.

Speaker 6

The next question comes from Antoine Gara with Wedbush Securities. Please go ahead.

Speaker 10

Oh, thanks for taking my question and congratulations on dual listing here. Just had a question, you know, high level. There was a recent paper published in Nature Communications about a month ago that demonstrated that quantum state teleportation between physically separate independent photonic emitters is seemingly achievable at, you know, pretty significant fidelity levels over, you know, meaningful distances. I think the demonstration was done over a 270-meter link. You know, I'm just wondering can you speak to this breakthrough? You know, what does it mean from a Xanadu perspective? You know, what does it mean for the broader space or the you know, photonics modality? Thank you.

Speaker 1

It was a great result, great paper. More and more stuff like that is coming out just in general. The difference there was really showing it in terms of a communication or an application, so in this case, teleportation. You know, our Aurora paper that was published in Nature over a year ago now really is much more You know, their paper was good, but ours is, just to be frank, much more technologically accomplished and more complex and achieves a lot more because you're now networking quantum computers or server racks together. For instance, we could have done an application like teleportation and kinda showed something similar, but it was really more about showing how to connect them, knowing that we can do algorithms later on. Overall, a great academic result.

Speaker 1

It was a nice result, that paper you mentioned. You know, the Aurora paper was a far more complex networking system in terms of quantum computing. It had like 35 chips over 13 kilometers, 4 server racks. Very complicated and, you know, the team's very proud of that achievement.

Speaker 10

Great. Thank you.

Speaker 1

Thanks.

Speaker 6

The next question comes from Tyler Anderson with Craig-Hallum Capital Group. Please go ahead. Mr. Anderson, your line is open. Please go ahead with your question.

Speaker 9

Sorry about that. I'm on 2 calls. Thank you guys for taking my questions. This is Tyler Anderson on for Richard Shannon. I was just wondering if you could highlight the milestones that you expect to reach within Optimism and what success with that program looks like and the timeline of it.

Speaker 1

Thank you for that. We're not at liberty at this moment to sort of talk about the milestones for Project OPTIMISM. The government won't let us because we're in final negotiations. Perhaps what I can say, though, is they mirror very, very well what we need to do for our roadmap that, you know, high-level roadmap that we mentioned, going through the qubit factory fault tolerance and ultimately a large-scale quantum computer. They map very, very nicely to that, and that's essentially what our funding will be used for from the Canadian government. You know, more details on our roadmap.

Speaker 1

Stay tuned, as Michael mentioned, for, you know, towards the end of this summer, we'll have a detailed roadmap. Hopefully, OPTIMISM, fingers crossed, is announced by then, and we'll be able to give you a better answer.

Speaker 9

Okay. Thank you. Could you just describe how your customer conversations have changed since you've become public?

Speaker 1

That's a good question. Well, I can definitely tell you our hiring's gone better now that we're public. I would say, slightly more traction, but I would say we, you know, the people that wanna get in quantum, which is now becoming more and more of the large corporations, are already in there, you know, with discussions. We've known them for many years. We're forming more all the time. Nothing's really changed there. It's hard to read. I would guess they would probably feel more comfortable perhaps now that we're kind of a long-term company by going public rather than sort of a scrappy startup. We haven't really seen any more meaningful. You know, nothing's really changed given the fact that they're like, "Oh, now you're public.

Speaker 1

Now we're gonna work with you." That definitely is not the case. We are one of the leaders in this space, along with other great companies, and they've always seen that, particularly with PennyLane and our photonic hardware. Maybe we can update you at our next earnings, once we've been public for a little bit longer and see if it's changed, but we've already had a good reputation. I don't think it's really changed too much.

Speaker 9

Got it. Well, people are everything. There's not many people in quantum, so that's great to hear.

Speaker 1

Exactly. Yeah, that's right.

Speaker 9

Thank you.

Speaker 1

Thank you.

Speaker 6

This concludes our question and answer session. I would like to turn the conference back over to Dr. Christian Weedbrook for any closing remarks.

Speaker 1

Well, thank you for that. I just want to say thank you to everyone on the call today. We appreciate everyone's time and questions, and looking forward to continuing the relationship beyond today. Thank you, everyone.

Speaker 6

The conference has now concluded. Thank you for attending today's presentation. You may now disconnect.

Powered by Quartr