theCUBE + NYSE Wired: The AI Factory - Data Center of the Future | Pete Shadbolt, PsiQuantum

theCUBE.net

- Links
  
  theCUBE Network SiliconANGLE Media
info Help

Pete Shadbolt, PsiQuantum

Exploring the Advancements and Impact of Quantum Computing on AI and Supercomputing

In this insightful discussion, Pete Shadbolt, co-founder and chief scientific officer of PsiQuantum, joins John Furrier of theCUBE to delve into the evolving landscape of quantum computing. As the co-founder of a leading quantum computing company, Shadbolt shares their perspectives from PsiQuantum's significant progress, including recent billion-dollar funding. This session is part of theCUBE's program from the New York Stock Exchange in collaboration with the Wired Community.

Shadbolt brings a wealth of expertise in quantum computing, steering PsiQuantum to the forefront of technological innovation. The conversation covers a broad spectrum of topics including the current state of quantum computing, its commercial applications in sectors such as material science and drug discovery, and the pivotal role of standards in advancing this nascent technology. Hosted by John Furrier of theCUBE and enhanced by insights from theCUBE Research team, this segment is part of ongoing discussions about technology's impact on society and economy.

Key takeaways from the discussion include Shadbolt's outlook on the rapid advancement of quantum computing, with PsiQuantum spearheading efforts to build scalable and fault-tolerant quantum computers. According to Shadbolt, while quantum computing is a few years behind AI in commercial deployment, its potential applications and implications are profound. The talk also explores the synergies between traditional semiconductor manufacturing and emerging quantum technologies, paving the way for revolutionary breakthroughs not only in computing but also in cryptographic challenges.

Share this session

Clips
More from theCUBE + NYSE Wired: The AI Factory - Data Center of the Future

Pete Shadbolt

Co-Founder & Chief Scientific Officer

PsiQuantum

play_circle_outline Introduction to theCUBE and the NYSE Wired Program focused on AI and quantum computing.

play_circle_outline Peter Shadbolt Highlights $1 Billion Funding and Evolution of Quantum Computing Towards Commercial Viability and Cybersecurity Transformations

play_circle_outline Unlocking Quantum Potential: Applications, Fault Tolerance, Semiconductor Manufacturing, and Open Source Collaboration in Quantum Computing

play_circle_outline Observations on unexpected interest in silicon photonics technology linked to quantum computing advancements.

Info
Transcript

Pete Shadbolt, PsiQuantum

Pete Shadbolt

Co-Founder & Chief Scientific Officer PsiQuantum

In this theCUBE + NYSE Wired segment from AI Factories – Data Centers of the Future, PsiQuantum co-founder and chief scientific officer Pete Shadbolt joins theCUBE’s John Furrier to chart the reality of quantum at AI scale. Shadbolt explains why the industry is entering a “phase transition” from labs to commercially-useful, fault-tolerant quantum computers that are “poised to break ground … within a matter of months”. He details PsiQuantum’s approach: leveraging high-volume semiconductor manufacturing to build silicon-photonics systems with early focus on che... Read more

explore Keep Exploring

What recent funding achievement has PsiQuantum accomplished, and who is their Co-Founder and Chief Scientific Officer? add

What is the current state and future outlook of supercomputers and quantum computing? add

What is the company's approach to building a quantum computer, and what applications do they anticipate for its use? add

What advancements in technology are being utilized in the development of quantum computers and how are they attracting interest from AI supercomputing? add

bolt Powered by CUBE AI

Pete Shadbolt, PsiQuantum

search

>> Hello, welcome back to theCUBE here at our New York Stock Exchange Studio as part of the NYSE and theCUBE Wired Program and Community. We've got a great series on the AI factor, the future of computing, the future of data centers, the future of large scale software systems that are rolling out, of course. Quantum is a big part of it. Peter Shadbolt here, Co-Founder and Chief Scientific officer of PsiQuantum, raised $1 billion in their recent funding and continue to push the envelope. Peter, thanks for coming in remote into our studio here in New York. You're actually where our other studio is in the Bay Area, so thanks for coming in.

Pete Shadbolt

>> Oh, thanks for having us. I really appreciate it.

>> Obviously the NVIDIAs and the open source area in large scale computing, GPUs, XPUs has really brought a lot of attention to quantum. And you go back a couple of years ago, quantum was, "When is it going to come out, who's got what?" You started to see real visibility into an accelerated push towards that next level. Some of the best companies now are working, writing algorithms, the NIST standards are aligning. You're starting to see real enthusiasm converting into confidence. Can you share quickly, you guys just raised $1 billion on your last financing, well-funded, going hard. Where are we on the reality? I mean, I know it's very nuanced, but share your opinion on what the market's doing right now. Clearly, it's not just being ready, people are doing stuff.

Pete Shadbolt

>> Yeah, I mean, I think firstly it's exciting that people are building big computers again. It was pretty unfashionable to build supercomputers 10 years ago, and currently it's very fashionable to build supercomputers for AI and so on. That's not the environment in which we're operating. And then this idea of a categorically new kind of supercomputer, a quantum computer has been around for decades. Academics have been working on it in universities for a long, long time. And then I think you're absolutely right that the field is now going through a phase transition where it's coming out of the research lab, people are starting to knock down some of the major technical milestones. People are starting to get serious about building really large scale, commercially useful machines. And we are poised to break ground on really, really big systems within a matter of months. I think it's a really exciting time for the field. Today, nobody has a useful quantum computer, and so in that sense, we're a few years behind AI, I would say, which is undoubtedly useful today. But we're moving at a very, very fast pace and really excited to get this stuff over the line.

>> Well, congratulations. And we love following the progress. It's funny if you even go back 12 months, right? "Oh yeah, quantum, that's 10 years away." And then it's five and insiders, "It's three," right?

Pete Shadbolt

>> Yeah.

>> And I smile because you never know the pace of the innovation the way the market is right now with computing. And I have to ask you too, because one of the things people talk about in the press, I just want to get it out there because it's talked about, is they call it the Y2K moment. There's a little doom and gloom. I don't mean to be negative and I'm not, but there's a little bit of doom and gloom around it's a Y2K potential extinction event if these keys are encrypted and harvested and decrypted, it could take down Bitcoin. That's kind of on the doom and gloom, but yet, there's a science side of it. There's things like how do you go from super computing today, which, we're seeing massive acceleration from even the old high-performance computing terminals that were moving inch by inch every year. Yeah, new thing, it only takes three weeks to do that wing design if you're Boeing, now it's faster. What's the bridge between the reality of the science and reality? Because there's a lot more than just quantum. There's fault tolerance, there's systems design, there's transitions, and then there's maybe pure play use cases. Let's talk about that reality and then we'll kind of get into what people are doing on pure quantum.

Pete Shadbolt

>> Yeah, our company is building a very large scale fault-tolerant quantum computer. We're doing it through leverage of high-volume chip manufacturing, so we build our wafers in a commercial semi-conductor fab. And we are going to use it primarily for commercial applications in chemistry, material science, drug discovery, fuels, catalysts, fertilizers, new semi-conductor manufacturing processes. We're going to use this machine to accelerate innovation at the foundations of our advanced society, and we think that's a commercially lucrative thing to do. You've also touched on cryptography and the potential for quantum computers to break current cryptographic methods. And that is not fantasy, that's a really serious future application of quantum computing. That is a double-edged sword. It's a powerful tool and it also comes with risks and challenges and all sorts of profound implications. I think the simple thing to say on that topic is that we expect the former set of commercial applications to come well before the cryptographically relevant systems are ready. And so my glib way of expressing this is that we expect to get rich before we become dangerous. There's a parallel universe in which nature conspired to force us to become dangerous before we get rich, and that would be unfortunate. Luckily, we don't live in that world.

>> Yeah, and that's a great comment. I want to highlight some of the things you mentioned around the forcing function. In all these markets, standards are huge, right? Can you comment on the role of standards? Because this is what can galvanize these big step function changes because this is again, going to be a significant architecture change. I mean, it's more power, horsepower for sure, lack of a better description. What's the role of standards? Because in all of these key market flexes of, go back 40 years in computing, the future computing was grounded in either academic research, government funding and standards, huge. Can you comment on the role of standards and how academic and then now commercial entrepreneurs can align and why that's important?

Pete Shadbolt

>> Yeah, so quantum computing is still very early. It's emerging out of research groups where people are doing stuff that has no standards at all, right? Like complete wild west physics research by a bunch of PhD students is where quantum computing starts out. But our approach has been to recognize that we can't do this on our own. That if we're going to get to these really big systems of millions of cubits, we need to leverage the $1 trillion and 50 years of work and alignment and standards that you described. We need to leverage that to our advantage. And so our whole approach really is to understand that we have to work with the semiconductor industry and work within the constraints of a big fab and an existing supply chain and contract manufacturing and so on and so on. We really need to find a way to fit quantum computing within those constraints rather than working against that existing capability by demanding atomic scale fabrication and crazy new materials and things that just kind of go completely against the grain of the momentum of the semiconductor industry. That was really our founding thesis and we've executed on that. We also just yesterday released our software tools called Construct. That's a software package for programming quantum computers, and that comes with a bunch of open source tooling for standardizing and capturing these quantum algorithms in a way that can be interchanged between teams. And so yeah, on the algorithms and software side, standardization, open source tools are really, really important in the same way that they have been for AI, things like PyTorch and Cuda and so on. We're working on very similar kind of landscape.

>> I love the fact that you're putting out that developer angle on that. I have to ask you, in all these big movements it's always like, and we've heard this from Steve Jobs, Sam Altman kind of refers it with OpenAI, in almost every major breakthrough there's always, "We meant it for that, but that actually happened." Are you seeing patterns? Because when you start throwing out the developers, you're going to see some usage because the alphas are coming in, it's a whole nother thing, it's a different game. But are there things that you see, either you expected or didn't expect? Because what will come out of that, I believe, will be something that will help you with your mission. If you're going to build these centers and continue to push the envelope, you're going to attract some serious people.

Pete Shadbolt

>> Yeah, one of the really cool things that has happened is that our approach to building the quantum computer is based on silicon photonics. That means silicon chips that move light instead of electricity. These chips were originally developed for telecom and networking applications, but we've needed to improve their performance well beyond the state of the art, so we've spent the last six years and more than $100 million making incredibly high performance photonics at GlobalFoundries Fab 8 in upstate New York, and that is suddenly becoming a topic of intense interest for the AI super computing people. So nothing to do with quantum, just people who are building regular AI supercomputers, they want extremely demanding specs on this advanced photonics stuff and it just so happens that we've got it. So that's one thing that it's still very early, we are not actually doing anything there yet, but it is sort of increasingly a frequent ask whether people can get access to that stuff. And so yeah, that's a byproduct of the main mission of the company, which is a huge, huge endeavor. But it's pretty exciting that we've already seen instances of these kind of unexpected byproducts of our work.

>> Well, great to have you on, Peter. My final question is more on what's next, and I want to kind of tease it out with an illustration. Five years ago, I went to Supercomputing, I think it was in Atlanta, I forget when it was. And that show started when I graduated college in 1988. It's like the oldest running show, and it was basically high-performance computing, terminals, time sharing, large-scale systems. I mean, state of the art. It just felt like a very boring, moving inch by inch, not a lot of needle-moving progress.

Pete Shadbolt

>> Yep.

>> I hear a little buzz, I go down there, I'm like, "Oh, this is just an HPC show." So I'm like, "Okay, I'm going to go walk to the restaurant." And down the hallway in the meeting area, I'm like, "Oh, AWS is here, NVIDIA's here?" I see signs, Nvidia, AWS and some other names, Broadcom. I'm like, "What the hell? What's going on here? This is not a cloud show." I walk in, I sneak into the room. It was hardcore pre-AI developers. I called Dave Vellante and I said, "Dave, this show's going to be huge. It's going to shift to be an AI show," because we just connected the dots on the spot. That as a preamble. We expect Supercomputing, maybe the show, but the category will evolve directly into quantum. Okay, so the question is what are some of the actors, who will be in that room? What's the equivalent tell sign in your opinion, of where that transition starts to connect? And again, this is a very simple illustration. AWS had cloud, they were doing nothing in AI. Now you look back and say, "It's obvious they were recruiting people." This is where now the game is going on. Quantum is going very quickly and supercomputing is basically the best advertising for Quantum. What's your take on what's that connective tissue look like? What's this hell sign? Where's the canary in the coal mine?

Pete Shadbolt

>> Yeah. Well, I think first of all, it's great to see Google, Amazon, IBM, Microsoft, Intel, all doing really serious work on quantum computing. And it speaks to the fact that it's appropriate to view quantum computing as the next frontier in a categorically new mode of computation. But I think that to your point, and I love your story of going to this event 10 years ago, when I came to Silicon Valley about eight years ago, people told me, "Don't say HPC, never say HPC. HPC is deeply unsexy, the whole idea of building big computers is very unfashionable. We're doing dating apps and dog walking and SaaS." And I think if you had told a lot of people in Silicon Valley eight, nine, ten years ago in a decade's time, the world's biggest company is going to be a supercomputer company that is predominantly doing what I think you can reasonably call scientific computing, right? The majority of people who run code on Nvidia supercomputers are scientists. It's not too much of a crazy stretch to say that Nvidia is a scientific computing company at this point. That would've blown people's minds, they would've said, "Absolutely not."

>> "You mean the gaming company? You mean the gaming company that does Bitcoin mining? What are you talking about? They just solved Matrix math, who cares?"

Pete Shadbolt

>> It was really exciting and I think it's part of a bigger trend where people are realizing that value is captured on the frontier and that it's no coincidence in my opinion, that many of our biggest and most durably valuable companies today are right out on the frontier of extremely hard technology. TSMC, ASML, SpaceX, Nvidia, I think that's actually a testament to what we're doing here. I think it's really exciting and obviously, we hope to one day earn our name in that list of companies.

>> Well, certainly we will definitely continue this conversation. You are on the frontier, value creation, value capture and contribution, the open source. Again, this is going to look a lot similar but be different than what we've seen. Major step function, change, major innovation. Again, we're going to look back and say, "Quantum? That was fantasy. That was a cybersecurity nightmare." No, no, no, it's going to be something bigger. And congratulations on the funding.

Pete Shadbolt

>> Thank you.

>> I'm looking forward to seeing how all these AI factories change the software equation, and AI is just a dream scenario, too. Software is going to change too, so great stuff. Thanks for coming on.

Pete Shadbolt

>> I really, really appreciate it. And great to see you again, John. Thanks for having us.

>> Okay, I'm John Furrier with theCUBE here in the New York Stock Exchange CUBE Studio East Coast. Of course, we've got our Palo Alto connecting Wall Street and Silicon Valley, tech and money together. The money's on the frontier, the work's got to get done. That's where the value capture will start. And of course, we're doing our part as part of the NYSE and theCUBE. The NYSE Wired Community, check it out, fast growing. I'm John Furrier, thanks for watching.