Thank you for joining us for today's live webinar, IonQ's Path to Large-Scale Fault-Tolerant Quantum Computing. We really appreciate you taking time out of your day to listen to us live. I'm Trevor Chaloux, Senior Director of Product Marketing here at IonQ, and I will be kicking us off. Before we get started, here are a few housekeeping items. Time permitting, we're going to take a few questions at the end of the webinar. If you have questions during our session today, please use the Q&A option on the Zoom console to ask your question. We will address some of those at the end of the session, time permitting. All participants will be automatically muted for the duration of this webinar by default. Please use the Q&A option, not the raise hand option, as we won't be able to address any raised hands. At the end of this session, you will get a short two to three-minute survey. We greatly appreciate your feedback. A recording of this webinar will be available after the presentation on our website, and registrants will be notified when it is available. Today's presentation contains forward-looking statements. We advise you to please review this language, which will be posted on the website, and to review similar cautionary notes, especially in advance of making an investment decision in IonQ. With that, I will hand it over to IonQ CEO, Niccolo de Masi. Thanks, Trevor, and a warm welcome to today's live viewers from across the globe. We have an exciting 45 minutes ahead. IonQ was founded with the goal of developing and delivering the world's first large-scale fault-tolerant quantum computer. Today, we are excited to share a significant update on progress towards that goal: the intent to acquire Oxford Ionics. Together with the recently closed acquisition of Lightsynq and IonQ's existing technology, Oxford Ionics' technologies is expected to accelerate our development roadmap. Peter Chapman, IonQ's Executive Chair, laid out IonQ's position on the value of near-term quantum computers in our 2023 letter to stockholders. The near-term commercial era in quantum computing, typically 50-1,000 physical qubits, does not yet need fault tolerance. At IonQ, due to the very high quality of our qubits and systems, this number of qubits is more than enough to perform useful computations. This morning's announcement of AstraZeneca is a prime example of our machine's capabilities with only 36 qubits. The hardware goals for this era are to unlock access to near-term commercially valuable applications and get to very low error levels on physical qubits to limit the amount of error correction needed later. With IonQ and Oxford Ionics technology, we achieve 99.99%, or four nines, average fidelity on our two-qubit gates. This two-qubit gate fidelity is best in class commercially and exceeds the highest fidelity achieved using any qubit, physical or logical, across the industry. IonQ's technology is built on exceptional quality physical qubits. With this, we can build the best logical qubits in the industry and scale to much larger numbers of logical qubits and faster because we have fewer errors to correct. At the end of the day, error correction and logical qubits are just tools for achieving higher fidelity or lower noise during computation. The full value from quantum computing comes from useful applications and large-scale usage, and that depends on three simple things. Firstly, how many qubits are available for computation. Secondly, how much noise can they tolerate. Thirdly, how accurately and quickly do applications run. As we will show here today, we believe IonQ currently does lead and expect IonQ to continue to lead the market on quantity, quality, and time-to-quality solutions for many years to come. We, as such, expect to solve an unparalleled number of valuable enterprise applications each year. IonQ remains firmly committed to enabling our customers and unlocking value from quantum computing investments now in the initial commercial era and increasing this value exponentially every year. Since becoming the first publicly traded pure-play quantum computing company in the world in 2021, we have built a commanding lead within the industry based on our strong financials, technology, and focus on commercialization. Our technology has been pioneering for over three decades, going all the way back to the first quantum gate operation demonstrated by IonQ Co-Founder, Dr. Chris Monroe, at NIST in the 1990s. Built on that pioneering foundation, IonQ's quantum computers have consistently led the industry in performance and scale. Thanks to our focus on innovation, we have deepened our technological moat with nearly 1,000 patents that span all areas of our core technology. Our leadership in the commercialization of our technology has afforded us a unique position and the ability to learn directly from customers about the current and future needs within the quantum market. This is true for both our pioneering quantum networking and computing roadmaps. Our global innovation in both of these key quantum ecosystems provides an enduring and compelling decision for our customers to work with IonQ. We have scaled our customer-facing teams globally with both sales and customer success organizations and have successfully engaged with enterprises, governments, research labs, and the academic world. Our world-class applications team has co-developed groundbreaking quantum applications to help our customers invent novel approaches to their hardest mission-critical problems. Today, we announced a 20x speed-up of key pharmaceutical computation work with AstraZeneca on our fourth-generation 36-qubit IonQ Forte Enterprise quantum computer. Our next-generation computer Tempo will deliver vastly more compute power. With future generations of systems, we will deliver billions of times more compute power. As I've said before, the era of narrow commercial advantage is indeed here, and the era of broad quantum advantage is just around the corner. With the opening of our Seattle facility in 2024, IonQ commissioned the first dedicated quantum computing manufacturing facility in the world, priming us to scale our production to meet current and future demand. Together, IonQ's leadership across financials, technology, and commercialization places us at the forefront of a rapidly evolving market and positions us to accelerate and extend our market leadership. In 2024 and 2025, we have grown our total addressable market, expanding further into the quantum networking market and accelerating our quantum computing roadmap through the acquisition of technologies and teams. Last week, we announced the completion of our acquisition of Lightsynq, a company founded by world-leading scientists from Harvard and Amazon AWS. Lightsynq's core technology is chip-based quantum memory. We will talk more about why you need quantum memory later in this presentation. This morning, we announced our intent to acquire Oxford Ionics. A spinout from Oxford University in the U.K., Oxford Ionics has developed a unique architecture with all-electronic gate control integrated on a chip. Oxford's integrated qubit control has crucial advantages for scaling qubit counts. We will spend most of today's webinar on how the combination of technology from IonQ, Lightsynq, and Oxford Ionics will create the winning quantum computer in each year and every era of quantum computing. IonQ's ecosystem stands for pioneering innovation and leadership from today through full fault tolerance, with the most compelling unit economics, groundbreaking performance from applications, and manufacturing simplicity and scalability. To discuss how these acquisitions will integrate into our existing architecture, it's my pleasure to hand off to Dr. Dean Kassmann, our SVP of Engineering and Technology. Thanks, Niccolo, and welcome, everyone. The acquisition of Lightsynq and the planned acquisition of Oxford Ionics represent a significant acceleration of our planned development work to realize our vision: to build the world's best quantum computers to solve the world's most impactful and complex problems. Now, the great thing is the component technology from both companies will fit right into our existing systems, architecture, and roadmap. IonQ has had the same basic design since we started. It's been a winning approach for almost 10 years. We're confident it's going to be a winning approach for the next 10 as well. At the heart of our strategy, we use ions as our qubits. We chose ions for a couple of really good reasons. First, every single ion is a perfect identical copy of every other one. You can think of it another way. Our manufacturing yield is 100%. They're manufactured perfectly by nature. The second thing we liked about ions is that they have an electrical charge, so it's easy to use that charge to hold and transport the qubit right where we want it. Trapped ions have the longest coherent times in the industry as well, thanks to their natural abilities and the fact that we can build traps around them. The whole idea of ion traps is not brand new either. It's based on decades of work in atomic clocks and leverages semiconductor manufacturing techniques. You do not have to take our word for it. It's pretty much undisputed, even by our competitors, that ions have the best native physical gate fidelity of any qubit that works today. The team at Oxford Ionics has actually set a world record on this, achieving a two-qubit gate fidelity of four nines, or 99.99%. Our whole approach has always been to use these super high-quality qubits and connect small groups of them within a single trap. Then we can connect those groups to other groups in other traps using photons or light. It is how today's most powerful supercomputers work. They use a distributed approach. This has been central to our scaling strategy from the very beginning. Ions give us a unique advantage here. Because we understand really well how photons and ions interact, it makes it easy to link them up. Now, our system design prioritizes making sure all of our qubits are highly connected. When they are, you get a ton of downstream benefits like compilation efficiencies and algorithmic design flexibility. Those things let us squeeze more performance out of our qubits that we have, even while they might still be a bit noisy. It also means we can use all kinds of error correction methods, both ones we all know about today as well as new ones that have not even been invented yet. Looking at the big picture, our long-term goal has always been to develop a fault-tolerant quantum computer. We believe our design will get us there with low overhead due to those extremely high-quality gates and high connectivity. When you combine that with constantly improving gate speeds and increasing parallel operations, we expect our approach will deliver practical solutions to big commercially valuable problems with practical time to solution. Finally, since we are a company that is 100% focused on quantum, we have always developed hardware and software together. By working on the full stack, we can maximize the performance of our systems while making sure they are easy for customers to access and use through our cloud platforms. We've been investing heavily in these areas for years to push our technology forward. Today's news is just the next step. These investments are going to create a massive leap in both our qubit count and how fast we can scale, bringing that horizon of our first commercially available interconnected quantum computer much, much closer. Now, the acquisitions of Lightsynq and Oxford Ionics not only bring new technologies and patents to IonQ, they also deepen our bench of scientific leadership. Dr. Chris Ballance from Oxford Ionics and Dr. Mihir Bhaskar from Lightsynq and their entire teams are leaders in the field. They bring tremendous expertise coupled with a proven entrepreneurial grit that helps us continue innovating at a high rate. We're thrilled to bring them on board and their talented teams. I'll have them introduce themselves in just a moment. Next, though, I want to touch on how the technologies, how these new technologies support our existing roadmap. First, Oxford Ionics' trap technology accelerates the qubit counts that we can put in a single trap. By moving from linear 1D architectures to a 2D qubit fabric, we expect to increase our trap densities 50-300x over time. That exciting scale increase comes with the same high performance, same parallel operations, and features like mid-circuit measurement that are expected in all of today's quantum systems and computers. Second, the Lightsynq memory-based photonic interconnects enable substantial improvements versus previous best-in-class approaches. Their technology increases the rate of photon-mediated ion-ion entanglement by up to 50x, which means higher throughput. I'll turn things over to Chris and Mihir in just a moment to explain more deeply how the technology works. First, though, I want to explain really where the rubber meets the road. What does all of this mean for our roadmap moving forward? Now, historically, we've always used a metric called AQ to mark our progress. For example, our current Forte Enterprise system is at AQ 36 with 36 physical qubits. With today's announcement, we're making a change. We're no longer going to anchor our roadmap against AQ. Moving forward, we'll be talking about both physical and logical qubits. Don't worry, when it comes to benchmarking, we're still on track to introduce our new approach later this year when we hit AQ 64, just like we talked about on the last earnings call. Now, with the combined strength of IonQ, Lightsynq, and Oxford Ionics, we're now projecting that we'll have 10,000 physical qubits on a single chip by 2027. A year later, in 2028, we expect to have two of these chips interconnected for a total of 20,000 physical qubits in one system. Now, our path to getting these big numbers has always been about printing qubits on chips without needing anything exotic. The chips do not need to be chilled to absolute zero. We do not need some new yet-to-be-discovered material. We just use relatively standard off-the-shelf technology that many fab houses already have today. The technology from Lightsynq is what allows us to connect these chips together, letting us run computations across multiple quantum processors at once. The connections are made with standard fiber optics, the kind that you would find in data centers today. A good way to think about the acquisition of Lightsynq is to compare it to when NVIDIA bought Mellanox. That move helped NVIDIA go from building single GPUs for individual PCs to creating entire data centers full of interconnected GPUs for AI. We're doing something very similar for quantum. Now, just two years after that, our architecture, which scales up incredibly fast by connecting these 2D chips, will allow us to build a system with over two million physical qubits. Those two million physical qubits will translate up to 80,000 logical qubits. That is leagues ahead of where our competitors are projecting, especially when you consider some of their approaches have not even run a real-world application yet. The values are based off of resource estimates using the latest error correction codes. Importantly, by 2030, we expect these logical qubits to be incredibly accurate, capable of reaching logical error rates of better than one part in a trillion. That is what you need to unlock the most powerful fault-tolerant applications. Also, the flexibility of our designs means we can tune and update the error correction codes that we use over time. What does this all mean? It means we're likely to have by far the most logical qubits in every era and lowest manufacturing costs for commercial systems. Our economics are already compelling today, and they're going to get exponentially better with every new generation we ship. We're really looking forward to unpacking this roadmap in more detail over the coming months as we integrate the Lightsynq team and technology and as we finalize the acquisition of Oxford Ionics and begin to integrate their powerful technology. Now, to learn a little bit more about the Oxford Ionics technology and how it will enable scale and performance acceleration, I want to please introduce Dr. Chris Ballance, the CEO of Oxford Ionics. Thanks, Dean. It's a pleasure to be joining you and the IonQ team on today's call. By way of introduction, I've worked at the forefront of the quantum computing industry since 2009. I did my PhD in trapped ion quantum computing at the University of Oxford, where I set multiple world records in quantum computing performance, including the highest performance quantum logic gates and the longest memory coherence time. In 2019, I founded Oxford Ionics alongside my Co-Founder, Tom Harty, and I currently serve as its CEO. As you mentioned earlier, we share a common vision to bring the most powerful and useful quantum computers to the market. At Oxford Ionics, we've spent the last five years tackling what we believe are the major bottlenecks to scaling this technology head-on. Firstly, by pushing the limits of qubit performance. Secondly, by developing a unique integrated qubit control that allows us to control ion qubits using electronic chips and then mass manufacture them through the might of the existing semiconductor industry. As of today, Oxford Ionics' technology has demonstrated the world's highest gate fidelities, with two-qubit gate fidelities at a whopping four nines, 99.99%. This means that the fidelity of our physical qubits is better than the best logical qubits ever demonstrated. As part of IonQ, we'll have a chance to significantly accelerate the development of our roadmap and make our shared vision a reality. Let's start with an important question. Why do we believe qubit performance is so critical? On one hand, it allows us to tackle commercially relevant use cases much earlier than anticipated, unlocking significant business value much sooner before we even need to turn on error correction. On the other hand, the moment we do turn on error correction, the overheads we require are dramatically reduced, making our systems overall less complex and ultimately requiring fewer resources to extract value. Better qubit performance gives you a near-term win and a long-term win. This was always IonQ's approach, and Oxford Ionics' technology accelerates that plan. Aside from benefiting from the perfect identical nature of ions, our unique integrated all-electronic qubit control technology enables us to engineer the performance of our qubits. How does that work? We took our inspiration from classical computer chips, the intrinsic simplicity that is at the core of their miniaturization and ever-increasing power. What we figured out is that the only way to achieve this simplicity and scale within trapped ions is to integrate everything needed to both trap and control the qubits into a classical chip built in a standard fab. Not only would this leverage existing technology, it would also allow us to leverage the power of the well-established trillion-dollar semiconductor industry to manufacture and scale our quantum chips. The architecture leverages electronics to control qubits through a technology called electronic qubit control. At the core of this architecture is an integrated antenna built into a silicon chip. When an oscillating current is applied to that antenna, the qubits experience oscillating magnetic fields that drive the quantum gates, allowing us to implement all quantum operations using electronics instead of the conventional lasers. There are several crucial advantages to this approach. First, we can scale our architecture more easily. Electronics are easier to chip integrate than lasers, allowing us to distribute qubits on 2D grids with the ability to easily control each qubit individually. We can also run quantum operations in parallel across larger devices, reducing the runtime and resources required. Finally, and this is an important one, we are the only established technology to build these chips, meaning we can leverage the tried-and-tested standard semiconductor manufacturing process, reducing both time to market and time to solution. Being able to leverage existing semiconductor foundries and supply chains is a massive advantage. As we scale quantum computers to ever larger and more complex machines, it is crucial to be able to do so with high yield and at as low a cost as possible. Dean, you said it well earlier. I simply do not think that many of the other technologies out there are cost-effective or production-ready. Our unique technology gives us an edge. Our architecture is incredibly scalable. Again, following in the footsteps of the classical computing industry, we've always engineered our systems so that small individual unit cells can be tiled across to create larger and larger devices. This gives us the ability to increase capabilities by replicating, not reinventing designs. We have a clear path to applying this to systems with tens of thousands of qubits in a single chip, and we've been actively working on our 256-qubit quantum processor units to date. We're incredibly excited to leverage IonQ's expertise in accelerating this technology board. Our technology allows us to scale devices to millions of qubits by building bigger and bigger chips. What is more, these chips can be networked by photonic interconnects to allow for distributed computing. This is where the technology from Lightsynq becomes critical to IonQ's rapid scaling strategy. To discuss that technology, I'll hand off to Dr. Mihir Bhaskar, Senior Director of Quantum Interconnects at IonQ. Thank you, Chris. As a brief introduction, I've been working in quantum technologies for over a decade, most recently as CEO and Co-Founder of Lightsynq Technologies. Before Lightsynq, I built and led AWS's Center for Quantum Networking. Prior to that, at Harvard, my co-founders and I built the first-ever quantum memory capable of speeding up a quantum network link. As Dean mentioned, photonic interconnects have been a linchpin of IonQ's architecture since the very beginning. Now, with the integration of Lightsynq's technology, the time horizon for commercially available distributed quantum computers exceeding 10,000 physical qubits has moved forward to 2028. The photonic interconnects we have brought to IonQ have three game-changing advantages. First, and most importantly, our interconnects leverage the world's best heralded quantum memories. These memories are based on photonic integrated circuits in diamond. Each nanoscale circuit element contains a quantum memory that can efficiently store quantum information carried by optical photons. I'll dive deeper on how this helps us achieve higher entanglement rates between multiple QPUs in a moment. A second key advantage of our platform is that our memories can be fabricated by the thousands using standard fabrication processes already widely in use at professional foundries. Just as Chris explained, this unlocks cost-effective scaling of our chips to support distributed architectures carrying millions of physical qubits across multiple QPUs by the end of the decade. Finally, photonic integrated circuits are important for everyone doing quantum computing, but they just aren't useful unless you can interface them efficiently with optical fibers. This is a major challenge across the photonics industry. Lightsynq has developed a proprietary fiber-to-chip packaging technology that reaches losses lower by about a factor of 10 than the industry standard. This low loss gives us an inherent advantage in attaining the most efficient and fastest possible networking speeds at scale. Let's take a look at how this all comes together. The previous best-in-class approach to photonic interconnects required us to detect entangled photons coming from two separate QPU nodes simultaneously. This means if a photon is lost en route from one of the QPUs to the detection hub, the entire connection fails, and the entire system has to reset and try again and again and again. This really throttles the overall networking speed. Lightsynq's quantum memories change everything. They completely eliminate the requirement for simultaneous photon arrival. Our memory can store a photon sent by one QPU while waiting to receive a photon from the second QPU, completing the link asynchronously. You can think of our memories as the quantum analog to network buffers that are found in virtually every classical network today. It is this buffering that makes our system resilient to unavoidable losses. When you plug in realistic values for these losses, the memory that Lightsynq has brought to IonQ can provide up to a 50x boost in the overall rate of entanglement between QPUs compared to the previous best-in-class approach. A key design choice from the very beginning is that these quantum memories do not require us to re-engineer the ion trap at all or redesign our architecture. The technology drops directly into IonQ's existing photonic interconnect architecture and works with our existing and planned ion traps. The speed boost that they provide brings forward the timeline for the first commercially available distributed quantum computer to 2028. I want to emphasize that this speed boost isn't just a theoretical projection. It's proven. It's based on over a decade of research that the Lightsynq co-founders and I carried out. In 2020, we actually showed that this works. We showed a quantum networking speed up of over 50x using our memory. In 2024, together with our colleagues at Harvard, we showed that our memory works with real telecommunications fibers like those deployed under the streets of Boston and in data centers around the globe. Our memories allow quantum computers to interconnect both inside the data center and outside into wide area networks as a key building block of the quantum internet. Today, I'm beyond excited to be a part of IonQ and have Oxford Ionics joining our company. We've known for a long time that trapped ions are the leading platform for scaling out QPUs using photonic interconnects. In fact, IonQ's Co-Founder, Professor Chris Monroe, practically invented the concept of networking QPUs over a decade ago. What most people don't realize is that over the past five years, the world record for high rate, high fidelity entanglement between QPUs has either been held by Chris Monroe himself or Chris Ballance and the Oxford Ionics team. That is what gets me so excited for the future of quantum computing at IonQ. The union of Lightsynq's team and technology with IonQ and Oxford Ionics represents a new best-in-class for quantum interconnects. Together, we're going to accelerate and grow our lead over the rest of the field. Ultimately, all of this underlying technology is designed with a singular focus in mind, which is to deliver systems and access to our customers which are capable of solving their hardest and highest value problems. To discuss how the application space for these larger fault-tolerant systems will evolve, I'll pass it over to Ariel Braunstein, SVP of Product and Applications at IonQ. Thank you, Mihir. IonQ is incredibly focused on delivering near-term value to our customers. Our partnership with Ansys is a great example. Ansys is a global leader in computer-aided engineering, and their simulation program products are used across the industry. For this collaboration, we developed a new algorithm that plugs directly into one of their leading products called LS-DYNA. We use a quantum computer to preprocess a 3D model with over 2.6 million vertices and 40 million edges. It's a large model. We then fed the quantum-optimized data back into LS-DYNA. The result was an improvement in simulation performance of up to 12% on their production software workflow. More importantly, it is the trend that appears in these results that matters. Per the slide on the screen, when comparing quantum to classic preprocessing with varying number of nodes, quantum shows an increase in simulation efficiency with every additional qubit. The hybrid workflow outperformed the current classical approach, even when given thousands of times fewer nodes to work with. These simulation speed-ups also translate into energy and cost savings for related simulation workloads. This is an immediate value that we can deliver today on Forte Enterprise, and we expect transformative results with larger systems. Similar implementations could unlock commercial value across many industries like automotive, aerospace, supply chain, manufacturing, financial services, and more. For another powerful example of near-term application, we turn to our collaboration with NVIDIA, AstraZeneca, and Amazon AWS. Together, we tackled one of the toughest R&D challenges around accelerating drug synthesis. This collaboration focused on building a quantum-accelerated computational chemistry workflow. In one of the largest quantum chemistry simulations ever executed in quantum hardware, we modeled a key step in a nickel-catalyzed chemical reaction used in the synthesis of small molecule drugs. The workflow integrated IonQ's Forte QPU, NVIDIA's CUDA-Q, and Amazon Braket. The resulting hybrid process demonstrated at least 20x speed-up in the time to solution compared to the previous state-of-the-art implementation. In an industry where development cycles can stretch over a decade and cost billions, these are meaningful results. Just note that this is again with only 36 qubits. For more details on the observed speed-ups, I highly encourage everybody to follow the QR code on the screen. As our next example, let's look at AI. As LLMs grow more powerful, their computational needs are skyrocketing. Classical hardware and energy infrastructure are starting to hit a limit, a wall, and IonQ technology can help push through these limits. We developed a hybrid fine-tuning framework that combines classical AI with quantum layers, and the results are really exciting. We took a standard sentence transformer like common AI tool we all use and paired it with a quantum circuit that acts as the classification head, giving the model new quantum capabilities. The result, when using common benchmarks, showed that this hybrid approach with as few as 18 high-quality qubits is already outperforming classical fine-tuning methods in accuracy. This architecture brings the best of both worlds. It allows us to repurpose extremely large, extremely expensive LLM models while training them for new tasks with minimal new data. Again, this can be done today, and the commercial impact will only grow with each system generation on our roadmap. The newly accelerated roadmap that we unveil today will have a direct impact on what will become possible over the next five years. What you see on this slide is only a small sample of the many categories of quantum applications that get unlocked at various stages along our roadmap. All of these applications unlock real commercial value over the next three years. In 2029, as the number of logical qubits increase and error rates continue to decrease, even more value gets unlocked. From computational chemistry in pharma to material design and manufacturing, medical imaging in healthcare, simulation in engineering, commercial value from these applications compound, and the market opportunity is vast. As an example, one exciting application gets unlocked in multiple stages. Catalyst redesign, as a use case, begins with around 2,000 logical qubits and touches so many industries, from oil refining to pharma, agriculture, and green energy. Today, the process of designing new catalysts is still largely trial and error as classical computers struggle to model the chemical interactions with precision. By 2028-2029, with about 2,000 logical qubits, we begin tackling a global challenge: designing a more energy-efficient fertilizer-making process, which currently accounts for almost two percent of global energy use. With about 3,000 logical qubits, we expect to be able to redesign existing catalysts to use cheaper, safer, and higher-performing materials, especially in pharma, refineries, and automotive. By 2029, with 4,000 logical qubits, we begin to engineer next-generation catalysts for carbon capture technology, vital for building a resilient planet. For our customers, this means faster design cycle, cheaper materials, less waste, greener processes, and most importantly, less dependence on precious material. Very valuable benefits. This all adds up to a very promising market opportunity for IonQ just from a single use case. New use cases will get unlocked on IonQ systems continuously, not only by IonQ, but by the entire quantum ecosystem. To close the presentation portion of the webinar, I will hand it back to Niccolo. Thanks, Ariel. Before I open this up to Q&A, I wanted to talk about the fault-tolerant quantum computing era. Each era we have discussed builds upon the prior one. Like building a house, one starts with a solid foundation as the first step. We began decades ago with ions because of the numerous technical advantages we discuss today. Widely known in the industry, trapped ions have led commercialization because of their amazingly low error rates. As we move towards the early commercial fault-tolerant era and ultimately very large-scale fault-tolerant systems, IonQ has established itself as the clear leader, allowing it to compete with the largest tech companies on earth and win. Let's open this up to some questions now for the team. Thank you to everybody who submitted questions throughout this presentation. We do not have time to get to all of them, but we have worked to try to synthesize them down into a couple of core themes that many of you are very curious about. First question, Dean, we will direct to you. The question is, many questions came in about what are the commonalities between Oxford Ionics and IonQ's technologies and how will these be integrated together? It's a great question. You know, there are a lot more similarities than differences between our two technologies, right? We obviously are both a trapped ion quantum computing platform, right? We rely on ions as our qubit source. We're both using barium in terms of a species for those qubits. We both have high native gate performance, high coherence times, right? Because of some of those similarities, that means we also have similar technologies for our qubit state preparation and measurement. That means our laser systems and everything are also the same. Both architectures have high connectivity in them. Our traps operate in similar vacuum environments. Our control electronics all have to both do the same kind of ion transport on comparable timescales. All in, there's actually a tremendous number of commonalities between the two technologies. Great. Thank you. Chris, maybe we will move to you next. As you know, we just touched on what are the commonalities here. A lot of folks are curious about the electronic qubit control that you discussed in your section of the presentation. Specifically, how is that different from the way that IonQ is controlling and operating on qubits today? The big difference is that electronics is far easier to integrate into chips than lasers. This means that the subsystems you need to trap and control the qubits can be integrated together into a single chip that can be built in a semiconductor fab. Ultimately, the semiconductor industry has been integrating electronics into chips for a tremendously long time. It's much easier to do that than highly precise coherent control lasers. What's interesting is that there's a really high barrier to entry to get electronic qubit control working. Once you have it working, there's a much, much lower cost of scaling. You're ultimately just building out classical chips on a standard semiconductor fab, and you can do these things without any new production facilities. We can really print these things off at scale. Great. Thank you. Let's move to final question for you, Mihir. So the intent to acquire Lightsynq was already announced, and there were discussions around how Lightsynq would integrate into IonQ's existing technology. With bringing Oxford Ionics on board as well, will the Lightsynq technology need to be further developed to support this combined architecture in the future? Thanks, Trevor. No, the quantum memory was designed with qubit modality interoperability in mind. That was actually a fundamental design choice we took when we founded Lightsynq. There are two key features of our technology which enable us to do this. First, we have a proven technology for photon wavelength conversion. This allows us to convert the wavelength of a single photon from one wavelength or color to another wavelength or color. We have shown that this works in the past, that we can make our memory compatible, for example, with the wavelengths that are already used in telecommunications infrastructure. Second, unlike competing memory technologies, our memories are very broadband. I mean that in the same way as we think of bandwidth and networking today. If you have a gigabit connection, you try and plug it into a megabit router, you're going to have a problem. Our memory, our interconnect, is much more broadband than either of the QPU technologies. It is compatible now and for many future generations to come. Together, these two features really allow us to seamlessly integrate IonQ and Oxford Ionics systems. As an added bonus, we're actually able to support cross-modal networking as well for use cases like we're delivering to the Air Force Research Lab. Great. Thank you. Lots of questions came in on the application section, Ariel. Maybe we'll take a pivot to you here. Specifically, there was a lot of applications listed on both the last slide and the slides in your section. Folks are curious, what are those applications and what is it that makes them great or ideal for how we've mapped them against our roadmap? Thank you. This has been the focus of our attention for the last four years, really focusing on understanding what makes great applications for quantum computers. At IonQ, we design the hardware in service of the applications, in service of the commercial value. I'd say at large, we can divide quantum application into these two large categories. One category is things that we can do better with quantum, and better can be faster, cheaper, higher precision, any axis, less energy consumption, so better. The second category is things that are just not possible in classical computing, where you enable something that was previously impossible to do. A few things that are important is that quantum applications never live by themselves. It's always part of a larger workflow that quantum needs to integrate with. Having domain expertise in both quantum and in the specific business, the domain that you're trying to innovate and the integration in that workflow is critical. Partnership, as you can see, we always emphasize the quality of our partnerships because those matter the most. Other things that matter is the best challenges that quantum can solve are those problems that grow exponentially in complexity. Therefore, classical computing can never catch up. Those are the best opportunity to start opening the door for quantum and start showing things that start with doing something better and end up being things that you could not do later on. Decided impact matter. Nobody wants to invest a large amount of efforts in building new infrastructure unless it is worth it at the end of the day. Finding those opportunities that have big returns at the end of them. I think that, as I said in the example about catalyst redesign, things that progress, things that scale as quantum gets larger and more powerful are better opportunities than things that just get unlocked and stay constant. These, I would say, are the best characteristics of what makes a good quantum application. Yeah, makes sense. Thank you, Ariel. Unsurprisingly, lots and lots of questions came in on the roadmap, Dean. There were some new numbers shared. Folks are noting that this is the first time IonQ has started talking about logical qubit targets as well as performance. I think there are other roadmaps that are out there from our competitive set, and people are just trying to digest. How does what we have shared today compare with the roadmaps that exist out in the market from other modalities? Right, what we talked about today, what I presented in terms of our technical capabilities and kind of numbers going forward is based off of our proposed thinking and integration plan over the next five years, right? With the Lightsynq technology, the Oxford Ionics technology, it's what gets us to 2 million qubits by the end of the decade with roughly 80,000 logical qubits, right? And that's based off of resource estimation techniques with the latest codes, right? To my knowledge, these are the highest numbers for both physical and logical qubits that have been announced by any commercial system, right? With any company in the world, right? IBM, for instance, their roadmap is trying to develop a 2,000-qubit quantum computer by 2033, right? That gives you a feel for what we're proposing and what we've kind of laid out compared to others. Great. You noted in your presentation, we will continue to unpack this roadmap in the coming months. For all of you watching live, please keep your eye on our website as we will continue to update with the roadmap that you saw today and follow-on information. Okay, I think we have time for maybe two more questions here. While we are on the roadmap theme, maybe, Chris, we will ask you a question. Specifically, you had a roadmap come out fairly recently at Oxford Ionics that talked about some targets. How does the acquisition from IonQ potentially impact that roadmap? Does it impact that roadmap? The combination of IonQ, Lightsynq, and Oxford Ionics technology significantly accelerates scaling and productization. This is not one plus one equals two or even one plus one equals three. This is one plus one equals 10. By tapping into IonQ's existing engineering and production capabilities, we believe we can not only deliver the previously announced roadmap targets, but significantly extend our roadmap going forward. In particular, the inclusion of quantum memory enables photonic networks that will drive scale sooner than we would have been able to do independently, all the way through to supporting the 2 million qubit target by 2030. Great. That's exciting, the extension of what you were building already. Great. All right. Why don't we do one last question? Mihir, folks, I think as you were answering your first questions, there were some follow-ups that came in. There's a lot of talk about how Lightsynq is going to improve upon the work that is best in class and the work that IonQ has done before. Can you talk a little bit about kind of where those improvements are coming from? What is it that this is bringing? Yeah. Here I'm referring to best in class or previous best in class as the academic work that's been going on. For example, Chris Monroe's lab at Duke, just a few months ago, his lab set the world record for high fidelity, high rate between trapped ions using a photonic interconnect. Actually, prior to that, since about 2020, the record had been held by Chris Ballance and the Oxford team. This type of work represents bleeding-edge research, but it still needs to be faster in order to connect QPUs so that we can actually use them in a distributed computing context. This is where the memory comes in. How does that work? The way that both Chris Monroe and Chris Ballance and everyone has been doing this is by having two QPUs emit entangled photons and by taking those two entangled photons and combining them at a detection hub. If and only if these photons arrive simultaneously, you have a successful generation of entanglement between the processors. If you have losses in your system, which you always do in any optical system, you just cannot get rid of all of the losses. You have to try and try again until you have that simultaneous arrival. This is where the memory comes in. If you replace that detection hub or you actually add to that detection hub a memory that's capable of storing a photon, so the first photon that arrives gets loaded in memory while you wait for the second photon, you have a big speed up in the rate of entanglement generation. That is how our solution kind of improves upon the previous best in class. Great. Thank you. I think with that, we will wrap up the webinar. Again, thank you to everybody who submitted questions. Unfortunately, we did not have the time to get to everything. We will try our best to follow up with additional resources to help answer what many of you are curious about. At the end of the day, this is a very exciting moment for IonQ. We appreciate you all taking the time out of your day to join us live. We will notify you when a recording of this is made available. Thank you again. Thank you, everybody.