IP Group Plc (IPO) Earnings Call Transcript & Summary

Good morning, and welcome to the IP Group Plc Deeptech Presentation. [Operator Instructions] Before we begin, I'd like to submit the following poll. I'd now like to hand you over to Mark Reilly, Managing Partner, Technology. Good morning, sir.

Thank you, Lilly. Good morning, everyone. Thank you for joining this IP Group webinar. This morning's webinar is an insights session. So this is all about how we're thinking about future technology and the areas that we're currently investing in the areas that we're looking to invest in new opportunities. And today, presented by the Deeptech team -- the IP Group Deeptech team, whose mission is to deliver value through growing innovative companies that enable them secure the digital economy, create new human capability and generate prosperity for all. Since we formed the partnership back in 2018, the Deeptech team has been doing this very successfully and has delivered a lot of success for the group. You hopefully will be familiar with the fact that we sold WaveOptics, an augmented reality company, to Snap for over $0.5 billion a couple of years ago. That was the deeptech team's exit and there's a list of others as well. In addition to that, we sold our Yoyo Wallet to Teya. We sold Processes Systems Engineering to Siemens, Re:infer last year to UiPath. And the team has also built a list of companies now that we invested in a very early stage from seed stage first investment that are now sort of $100 million-plus companies. So great sort of track record of success in that team. And we've honed our focus over the period of the team's existence to a set of core focus areas now which are summarized on this slide. First of all, the team is focused on applied AI. So looking for uses of AI, where artificial intelligence can be applied to data sets that are well suited to generating insight to finding insight in a data set that can be commercially valuable that can deliver a value proposition to an end customer. And we have lots of companies in the portfolio now that are doing that and have a value -- proven value proposition and are generating revenue. And the second area of focus is on the changing nature of our interaction with machines. We think this is a really interesting area of how we're moving away from using keyboards and using mouse and the interaction with the computer is much more like an interaction with a human being where we're using gesture and voice and expression to interact with computers. And those 2 sort of core evolutions of the way that we interact with compute and the way that we use computing power has an implication for the underlying infrastructure and our communications networks and the compute requirement that we have underlying and supporting those use cases. And so there are 2 other areas of investment to focus for the Deeptech team, and that is really the focus areas that we're talking about today. All of this demand created by artificial intelligence models and artificial intelligence algorithms on the compute power is ever growing first for a deeper and more capable computer that consumes a lot of energy, and that also plays us on a burden on our communications network, so we're trying to shift over more like a larger quantities of data. And today, we're going to talk about some of the innovations that are required in order to address that, have a more pressing need. And to do that, I will introduce you now to Dr. Lee Thornton, partner in the Deeptech team. Over to you, Lee.

Thanks, Mark. Thanks for the introduction. Yes, so I work alongside market IP Group is one of the partners in the Deeptech team. And we've organized 2 speakers today to provide some interesting insights into the different aspects of the challenges and opportunities that Mark outlined around data transition, data movement and data generation, some of biggest challenges within the world moving the greater adoption of AI in terms of our computing processes and computing power. The first speaker will focus more specifically on data transmission and distribution with the world now well into the data. So 10 of the 21 bytes of data being generated and stored in data centers and the devices and got new data often used in places where that data is not generated. So there's a huge transmission challenge, both in terms of network management and optimizing network infrastructure. And as Mark says, first to date remains unquestionable, and that's only going to get more problematic how we shift data around the Internet to the user and then back to the processor. So that will be a focus of the first speaker. And the second speaker, we'll share some of the challenges associated with our current computing paradigm. And the way we currently process is shifting from CPUs to GPUs and then further on to next-generation computing hardware, is, again, the demand for AI models and different forms of processing continues to increase and change we need to think of new solutions, especially in light of silicon and Moore's law potentially coming to an end and silicon nodes getting down to 4 nanometers, there's not much more space to go. So we need new paradigms and new ways to process data, generate insights and of course, address the energy challenge. There's only some energy we can consume with all of this while still being responsible custodians. So the first up, I'll turn straight to our first speaker. I could say this, we'll do a Q&A panel session at the end. So if you have any questions at any time, please put them in the Q&A box, but I will hold the questions until both speakers have presented. And then we'll come back with Mark and the 2 speakers where we'll answer some of those questions and have a bit of a debate, as I say, so please feel free to add your questions. So the first speaker we have is set Cizel. He's the machine learning lead at Deep Render. He will be giving his view on video compression technology, past, present in the future. So Cizel is originally from Slovenia. So has a PhD in Spring Theory from Oxford. And after is PhD was looking for careers with more concrete real-world impact was converted to be an AI compression enthusiastic by the team at Deep Render and now Cizel leading a 20-person strong team and the engineering department at brand, so delighted to have them here and to provide us insights. Fantastic. I said that's over to you.

Perfect. Thanks. Thanks, Lee. So yes, hi, everyone. I'm Seb and I have the machine learning lead here at Deep Render. So first of all, I'd like to thank the IP Group for organizing this a provide a number and facilitating this discussion on the future of the Internet and how AI can influence the future. So we at Deep Render, I'm very excited to be pushing the envelope of video content delivery by developing world's first entirely AI-based video compression method. But before we jump into AI, we can ask ourselves the fundamental question, why video? So video calls are a staple of our life. The pandemic really introduced the concept of interacting solely via video. And personally, having families scattered around the world, the challenges of video content delivery became very clear, very quickly through that rate, I think your frozen message. So I think we've all experienced this at various points over the last couple of years. So we interact a lot via video. When we zoom out from just videoconferencing, video data is actually dominates the data transmitted over the Internet by far. Up to 70% of all Internet traffic is actually video, and this is rapidly growing year-on-year. So this underscores the point that dealing with video data effectively at a large scale is essential for a stable growth of the Internet. And the goal of my talk today is to present the underlying technology that makes this possible and the challenges of the past technologies and what AI has to offer in the future. So let's move on to draw some to underscore some arguments why compression is necessary. So the key takeaway from this presentation or like one of the key takeaways from this presentation is that compression is actually everywhere. Compression facilitates most of the big consumption that we individual users consume on the daily basis. So you are most likely interacting with something that has a compression algorithm running in the background every single day, and it's most likely for hours. This is due to the fact that most data that we consume and we generate is video. So anything from streaming Netflix, video conferencing to house gaming this very webinar is backed by video. So the background of this webinar, there, your computer is churning away and running thousands of compression algorithms every single minute to ensure that all of the slides are actually arriving on time to your screen. So right now, it seems to be doing a good job. But if it starts falling, we'll all feel the failure very quickly. So on a more global scale, these algorithms have run trillions of times every single day. So compression really is the engine that underlies the data transfer over to the Internet. But why is the compression of wise compression and specifically video compression so necessary? So the fundamental underlying fact is that video files are very large to. Illustrate just how large they are, we can consider a simple example. Let's take 1 hour of raw uncompressed 4K video, which I mean a video that comes out of the camera. I suppose you want to stream a Netflix movie of -- that consist of 1 hour of raw 4K video. How much data is that actually? That actually turns out to be about 3 terabytes of data, if you just wanted to store it on your computer. So 3 terabytes is a lot of data. It definitely doesn't fit in my laptop. It actually sits on 5 or 6 sort of average laptops. So it's very hard to even keep this data in the some compress form on disk. But suppose you was to stream it in real time without any buffering and lagging, what would that require? Well, that requires an Internet connection of 7 gigabits per second to be able to stream reliably. This is actually a massive amount of bandwidth. Turns out that it's about 100x the average broadband in U.K. at this moment. And the same is not only felt by the end user, it's also felt by content distributors. So every single gigabyte that has to be transmitted incurs a cost that cost us the bandwidth cost. So content providers pay a few cents for every gigabyte that leaves their storage service. So if you multiply that by user basis of $200 million, suddenly, it's very easy to reach bandwidth costs, exceeding billions and billions of dollars. So that's what underlies the fundamental need for efficient compression for an efficient way to reduce these file sizes to be able to transmit them over the Internet. So let's see what the past solutions to the problem of reducing file sizes have been and what challenges they face. So in the past, a lot of algorithms have been developed to compress videos to be generally referred to as traditional compression. The traditional compression has iteratively evolved over the last 5 decades. And the key objective is to reduce the file sizes while preserving individual quality of the data. So this is common to all forms of compression. You want to preserve the fidelity of the original file while reducing the cloud size. However, despite traditional compression being able to keep up with the data demands of the past, it has some key drawbacks that hindered going into the future. So first of all, traditional compression algorithms are very complex and had designed and have been designed over the last 5 decades. This means that they're fairly inflexible when it comes to doing the new pervasive forms of data of the present and the future. So for every new data modality, traditional compression has to be hand-specialized to ensure the optimal compression and this takes time. It also needs specialized hardware. So there's a high probability that every single device with a screen that you own has a little chip inside. The sole purpose is to run traditional compression algorithms and encode and decode video on device. It doesn't do anything else. It just deals with video compression. And the problem with that is that suppose you want to update your compression algorithm. Well, the specialized hardware is tailor-made for one generation. So for a new generation, you first have to design new hardware. You have to wait for the users to actually get the hardware which means that the process of actually updating compression algorithms is very slow. And a corollary of that is that there is slow market penetration, a slow response to the key pricing challenges for the increased data consumption process. So what does this leave us today? So today, we're seeing 2 concerning trends. So the amount of Internet traffic per month is increasing exponentially. You're generating more and more data, and we are transmitting more and more data to that user. The reason for it is we're generating higher quality data and we are introducing new technologies and new industries that are very data intensive. However, for traditional compression, the relative compression improvement generation over generation has been slowing down, which means that it's not able to effectively compress the data that's being generated right now. And it will be even less equipped to compress the data that will be generated in the future. So this problem is not a hypothetical problem. So you may recall that during the pandemic, Netflix was actually forced have the streaming quality of the videos in Europe because the increased network demands that the pandemic posed threatened the existence of the European network altogether. So this underlies the message that we are at a very critical juncture right now. So our infrastructure is very strained and every single terabyte of new data that's generated is adding to this. And the traditional compression is unable to deal with it. So without a revolutionary solution to this problem, we could see the collapse of the Internet data infrastructure. So what could be a solution to this problem? So at Deep Render are very confident that the solution lies in AI, specifically in AI-based compression. So the key advantage of AI-based compression is that we can leverage the massive amounts of data to develop rapidly updated, highly specialized algorithms that are responsive to the new data that's being generated and allow it much, much better compression ratios relative to traditional compression. So we think that AI-based compression is the future. So why do we think that? Well, first of all, we can already show that it works. We have developed AI-based compression techniques that result 80% or better compression ratios or pervasive presential compression that's widely available today. And we developed this over the last 5 years. So compare 5 decades of development versus 5 years of development and the result is already 80% compression. The possibilities for the future are even more optimistic. It's one of the qualities of AI-based compression that is harder to achieve in traditional compression, we can optimize the quality for the human visual system. So we, for example, in this webinar, are able to provide more detail to the key features of video like face and even more the relevant features like the background. The ability to train and iterate on the algorithms quickly, it means that it's flexible and widely adoptable. And finally, we also have a rapidly growing hardware and software ecosystem powered by AI revolution, which means that our algorithms will get faster and better, simply by the fact of the growing adoption of neural accelerators by end users. So who can benefit from this? So we believe that the benefits are concentrated both in the end users and the content providers. From the end user perspective, it removes the bandwidth constraints leads to higher quality content and a leap in customer experience fundamentally unlocks new types of content that are very data-intensive, for example, cloud gaming and virtual reality. On the other hand, content providers, -- could benefit from billions of decrease -- from a massive decrease in network bandwidth costs, expanded market access and accelerated adoption of new technologies because they are not mandated to wait for the adoption of traditional compression algorithms. So we at Deep Render have an amazing team of people that has been pushing the boundaries of AI compression and has achieved several world-first in the area of AI-based compression. We have a world first demo of an end-to-end video chat that's entirely backed AI-based compression which just yesterday led us to when the Intel Ignite start-up competition, and we will be featured in the Intel Innovation Keynote today. So we believe that we have the tools to provide a solution for the Internet problems of the future. But the aim is not the technical improvement. It's basically about reimagining the fundamental engine to drive the Internet data transfer and thereby dramatically improving user experience, opening up new market and allowing Internet to grow sustainably. So if you're interested about the future of AI-based compression and about the demos of the world first AI-based compression apps that developed Deep Render, please contact us. Thank you.

Very insightful. As I mentioned, we'll take some questions at the end after the next talk. So please, for those in the Q&A, will now already at. At the next up, we have James Spall who is a co-founder of Lumai. Who give his thoughts on next-generation hardware for AI and the return of analogue processing. Interestingly, James first experience using it was to help identify the creation of top box from the large collaborate. So -- but his second experience was somewhat more esoteric as a data science internship, he used AI to help create custom dog food subscriptions probably not something we see at IP, but very interesting nonetheless. James is now just finishing his PhD from the University of Oxford, working on optical neural networks before he joins Lumai full time in the autumn. So James, over to you for your presentation. Thank you.

Great. Thanks, Lee. And yes, a big thank you to the whole IP Group team for the invitation to speak today. So as Lee mentioned, I'm 1 of the co-founders of Lumai. So we're a spinout from Oxford, creating the next generation of hardware for AI using optical computing. So today, I'm going to be talking about kind of the processing side of AI, kind of the really crucial kind of fundamental underlying underpinning aspects of AI is the processing speed. So we're looking at kind of the current trends and challenges and then looking at why I believe that analogue processing is the next step for AI hardware. . So we are kind of undoubtedly seeing a time when kind of both pervasiveness but also the capability of AI is just exploding, right? It's increasing some incredible rate of just one example here. I think it's kind of obligated to use ChatGPT as an example of the capability and invasiveness of AI. But it really is amazing kind of how quickly this has unlocked people's awareness of the capabilities of machine learning. I think it's one of the fastest adopted technologies on the planet with ChatGPT kind of 100 million users in 2 months. It's just incredible how quickly this is taking off. And there are plenty of other examples besides, as Lee mentioned, I've done everything from kind of fundamental particle theory all the way down to just can we create better dog food subscriptions for people. It's incredible that kind of the vast array of areas that this is going to change. But what really drives that capability is the raw compute speed, right? And by that, I mean, the amount of competition, the number of processes per minute per second, that you can perform. So in some given time. And every kind of big step change in the capability of AI we've seen has been associated to kind of a few orders of magnitude increase in the compute speed. And kind of one of the kind of major drivers in that increase in processing speed has been a change in philosophy of the hardware that we use to do AI. So we're no longer running these kind of machine learning models on general-purpose computers on CPUs on your laptop or so on, but moving towards more bespoke specialized hardware. So overwhelmingly, in the data center using GPUs, graphical processing units and including the name the graphical, these were kind of originally designed for high frame rate, better picture quality rendering and games and graphics, but they kind of ideally suited for AI in the sense that they're very good at doing operations in parallel and doing things that kind of at the same time in a very efficient parallel fashion. So they've been adopted for use in AI, we're going further still. So you come across term ASIC, application-specific integrated chip. The application specifically here being AI. And we're seeing companies like Google and Microsoft, meta, all the big players are developing their own chips that are optimized just for the certain aspects of their machine learning workloads that they need, right? So we have this trend of going towards more bespoke, more specialized hardware that's much faster and more efficient for AI. But the growth of AI is kind of putting such huge pressure on developing these new types of processes that are better and better suited to AI. And the issue is here, all of these processes so far, use digital computation, these digital arithmetic binary processes, turning everything into 0s and 1s. And that's really important kind of rewind 50, 60 years. That's really important in terms of the growth of computing as a whole being able to have very high precision, very flexible, programmable computing. But that's not particularly what AI needs, right? AI just to be able to do very specific arithmetic operations, can be relatively low precision and kind of this idea of precise digital computation for AI, it doesn't match very well. And actually, the need for energy efficiency and compute outweighs that need for programmability, flexibility and precision. That's why I see kind of the future direction of this is to go to even bespoke, even more specialized analogue processes to kind of match the physical characteristics of the neural network and the physical needs of AI. So why has this kind of not happened already? If that's the case, why are we not seeing analogue processes everywhere so far? And so I'm going to kind of break through the kind of 3 kind of key areas of why now, why the future of? AI hardware is analogue now? So there's kind of -- there's a need for it. There's a new market for it, and we have the capability now that we didn't before. So when I say there's a need, what I mean is there's kind of real fundamental limits and issues with digital computation, right? There's some real financial issues. And the first one is just kind of physics. So the way that we build our processes now based on kind of silicon chips is kind of approaching, as Lee mentioned, at start, is approaching its limits. The number of kind of transistors that you can fit onto a chip is basically how small make each transistor. Smaller you're transistor, the more you can squeeze in, the higher the density, and that's been doubling every 2 years or so. That's Moore's law as the transistor density has been doubling every 2 years. But if our transistors get to a point where they're so small, there are only few atoms across, which is exactly where we are now, you reach a hard stop, right? You reach a limit. It's incredible the fabrication technology has allowed us to get to this point where each transistor is only a dozen atoms also across, but if you get down to kind of a few items, there's no way you can push that further. And that's a real issue. And that kind of ties into then the real kind of fundamental issue to the data centers and compute providers are facing now is the energy consumption. So each one of those processes using a passing current through each transistor takes a tiny bit of energy. But if we're doing this on the order of trillions of transistors and trillions of operations, that adds up really quickly to kilowatts of power per processor, right? And we're not just using tens or hundreds of these processes in each day center, we're using tens of thousands of this. And so when you add up all of that energy consumption together, that's the real limiting factor to how much processing power you can squeeze into one data center. It's not footprint, it's not anything like that. It's the energy consumption. So there's kind of 2 sides of this. One is that's obviously terrible for the environment, right? There's a huge carbon footprint associated with these things. And it just simply sustainable from that point of view, but it's also unsustainable from this need to push the compute speed further. The way you get better AI and the way you increase the capability of AI as you increase the complete speed. But if you're fundamentally limited by the energy consumption and the hardware that you have, you just can't push that any further. So just kind of a few astonishing kind of facts I found about this. Kind of each data center is using megawatts of power, it's equivalent to running like the energy needs of a town or a small city just for 1 data center just for AI and kind of somewhere between 10% to 20% of global electricity is going to be going towards data centers by kind of best decade. This is a great one from the Republic of Ireland has many data centers. It's a large part of their economy and already 14% of their electricity just goes to running AI. It really is kind of incredible numbers. So on that one hand, digital electronics is putting huge strain on our infrastructure. There's a real push for alternatives. But then at the same time, is there a kind of a market -- is there a space for specialized bespoke hardware like analogue processing to actually play a role in AI? And the answer is yes, because of kind of 2 factors. One is the need of AI is, really simple right? The kind of the underlying operations that are being performed is just really simple arithmetic. It's multiplication addition, that's about it, right, in the form of matrix multiplication. And so having kind of a niche processor that's only doing a few operations, it's fine for AI because you only need to do a few operations. And it turns out that analogue processes can do arithmetic fantastically well in a very efficient, very parallel fashion. And so it maps really nicely onto to performing machine learning and performing AI. The other aspect is kind of whether you actually have enough need or enough kind of total processing power required for such a niche bespoke processor. And if you kind of had just general or independent devices, as I kind of touched on, it doesn't make sense or it's very challenging to have a bespoke processor in individual devices and individual consumers' hands. But we're seeing more and more of this growth of cloud compute, right? This idea that you don't do your processing locally on your device. Do you send it via the Internet, via the connectivity that we know all of our devices now have, and it all gets co-located into a single data center. And so that kind of -- that growth of the cloud compute has provided this perfect market. There's perfect opportunity to house these thousands of coprocessors and specialized hardware for AI that we haven't had previously over the past few decades. And that's exactly why the likes of the names I mentioned, Amazon, Microsoft and so on are developing their own chips because there's such a huge need for all of this processing all located in one space that it unlocks that ability to fully utilized bespoke and specialized hardware. So we kind of -- we've seen the need the decline of digital electronics. There's a huge market from it for AI in the cloud. But do we have the capability? Do we have the kind of the technology to really introduce analogue processes? And so this is where I'm going to introduce optical computing. This is what Lumai are working on is this idea of using optics to perform computation and not digital electronics. And so it's exactly the same kind of benefits and ideas that we have in the communications industry. You no longer transfer and move data around on [ coax ] cables, you use fiber optics, right? And exactly the same benefits in that domain, you get with processing as well. So on the one hand, it is an analogue computation. And as we've seen it maps perfectly into the processing and the arithmetic needed for AI. And at the same time, you get huge advantages in terms of throughput and latency. So the clock speed of optics is much faster. You can use things like multiple wavelengths that you just don't have in the digital domain to get much higher throughput, much lower latency. And what's really critical is the power consumption, right? So when you're into the optical domain, you don't have issues of heating like you do in digital electronics. There's no passing a current through resistor and getting heating things like that. You don't have energy use when you pass a beam through a lens or reflect a laser from a mirror or something like that. And so that's really important is the energy efficiency of these analogue optical processes can be orders of magnitude larger. So to kind of summarize that, we see kind of the limits of the current hardware has provided this great opportunity, right? Kind of there's this brick wall that Moore's Law is about to hit in terms of transistor density and energy consumption. And there's a real need kind of moving forward and looking over the next few decades, what is our computational and architecture going to look like? And I don't think it's just digital electronics. Then at the same time, the growth of AI and the pervasiveness of AI and the move towards the cloud and the data center has provided this perfect market of needing bespoke specialized hardware that can do certain operations extremely well and kind of optical computing and other analogue devices that especially optical is going to be a really important aspect of the next generation of computing and filling that opportunity in that market. But there are other challenges, right? It's not all just about processing. There are other considerations that we need to address. So one of them we've mentioned is networking, right, data to and from the cloud. So -- so kind of touched on this, the importance of just getting that data around and moving that data is a huge challenge. And as we move towards these kind of ideas of cloud compute and data centers more and more, then that networking becomes a real pretty important aspect. The other aspect is within the data center, as I mentioned, we're not using a handful of processes. We're using tens of thousands of these devices. And so kind of scaling that number of devices in a sustainable way and how you interconnect to those different devices is very important. And finally, as we do move to these concepts of kind of bespoke specialized hardware analog devices, digital devices, optical electronic, kind of officially connecting and converting between those different processes is really important as well. And that's also part of what we're working on at Lumai is how do you integrate these new devices seamlessly with the rest of the compute stack. So I'll finish there. I think that's what we've got time for. Thank you very much for your attention. If you have any questions, please do reach out from me an e-mail or visit our website.

Fantastic. Thanks, James. Could I just ask Mark and Seb to come back is fantastic. We'll take some questions and have a bit of a discussion. So again, you can ask questions I can already see come in on the Q&A box on your screen to the right, and I will take the relevant questions and post into the group. So I think the first one there is probably quite specific to use. Are there any solutions in the pipe to filter on the video image person and the full brand of the video call while effectively filtering out other people in the background? Do you have any comments on that? Does that make sense?

Sure. Yes. So there are definitely solutions like that, that you're thinking about. So the advantage of AI is that you can do this end to end so, you can basically learn a lot of this just by showing the model, the data that you wanted to perform well. So -- and showing them all what it needs to focus on. So if you are able to convey that information to the model, focus on faces, or details, you can actually do the same end to end. So this is definitely an approach that we're thinking about.

Okay. Question that often comes up about AI regulation. And do you think AI needs to be regulated? I mean people have views on this, but any particular points anybody wants to raise anything to add to that debate?

Yes. I mean the high level, the answer is yes, right? It does have potential to be very dangerous, and that's something we ought to do something about -- I mean it's a big question and it's a bit like -- so the Internet be regulated. Yes, it has a potential to be dangerous, but it's not an easy thing to do, and there are dark corners of the Internet that's still very sorry dangerous today. It's good to see the U.K. make sort of proactive efforts on this. We've got the U.K. AI Safety Summit coming up in November. And a chat with Matthew Clifford has been appointed as the prime ministers invite to that summit, which is a good move. I think he's a very smart guy and has very good kind of thinking on this topic. But yes, it's a big challenge. And we will -- it will continue to, of course, create an opportunity no matter what regulation we apply. .

James, do you have any view on...

I mean, yes, I think it's kind of -- it's quite hard to think about AI safety as just kind of one topic because there are so many different aspects to it, right? This is a kind of misinformation side of it, ChatGPT telling you things that aren't true and can that be used to kind of do harm in various different ways. Then there's also kind of quite a lot of people talk about the kind of existential threat that we kind of see from AI. And that's a very different thing, right? There's kind of -- there's the immediate issue of misinformation and so on and a much longer term if we keep pushing AI as we are, does it at some point, get to a level of human intelligence and all sorts of those kind of conflict. So there's a huge range of things and they all need to be considered in different ways, right? You should -- don't need to regulate AI safety from the point of view of existential threat right now as we sold more and more with the miscommunication -- the misinformation.

Yes. Interesting. Seb, any view on that?

I think we've James here. I think there's a spectrum between sort of more hypothetical threats and more concrete threats from AI. I think inevitably, the regulation of AI will have to take some shape or form, but it seems to me that it's more fruitful to focus on the concrete, concrete issues.

Yes. Can I ask Seb and James, what your view is on the existential? I mean clearly, we're a long way from the illusion of intelligence that we're seeing in AI today. But at the same time, I'm not able to identify any kind of fundamental physical barrier to a computer becoming sufficiently intelligent that it could sort of out smarter and provide that existential threat. How do you feel about that? Is that...

I think as people's concern is that at some point, it just falls off a cliff, right? You don't see it until it's too late broadly. I think that's been a scary thing. That's what people are concerned about, right? It's kind of -- it doesn't just kind of incrementally get towards human-level intelligence. It will suddenly appear. And so it's kind of preempting that and getting the regulation in and the safety nets in ahead of time. Before that happens, that's really important. And that's, I think, a big -- a very growing field of research is just on the AI safety. How do you develop these neural networks and these AI models in a safe manner from the fundamentals rather than trying to bolt-on safety features after the fact. So it's a very interesting area of research, yes.

Yes, exactly. I think especially with the met with the large language models, we are all fascinated by the fact that we're able to interact with what is essentially a computer in a conversational manner. And language was one of the big frontiers of intelligence. So just thinking back to the Turing test. Language is something intrinsically so human that when a machine started seemingly being able to use language. This was a massive paradigm shift for most of us. So I do think that research in that area is important. I do think that sometimes conversations get overly focused on some hypotheticals and not sufficiently focused on the concrete issues that are already a problem. But that's why basically we need to do a lot of research or people working on large hybrid models, we need to do research, both in safety right now and safety in the future.

Just changing types slightly, you both talked a bit about the technologies in your companies. What do you think -- generally do you have any view on what else needs to develop outside of your specific areas? What are the technologies need to be developed to see really widespread adoption AI models both centrally and at the edge? What else needs to happen if you think do you think for what you guys do to come reality? I'll put that one to James first.

Yes. Well, so kind of as I mentioned, a huge part of the kind of the interconnection of devices, right? So the development of the infrastructure and the protocols of how do you move data around efficiently. So is that kind of touched a lot of it on the kind of compression side? I think that's kind of a large part of it is moving to the data, but also the kind of the compression of the AI models themselves, right? So ChatGPT 3 is 175 billion parameters or so on. By making the number of parameters bigger the model is bigger, the AI gets better, but then you have all these constraints on your hardware. So getting the same level of kind of AI ability shrunk down into smaller compressed models will be really important. So those are kind of the 2 things for me is the networking and the interconnection of devices and the compression of the actual models and the actual algorithms themselves.

Seb, do you have any comments on I mean is there any incompatibility issue with compression in AI? Or is it the 2 things just go hand-in-hand together, one doesn't negatively impact the other or anything? .

No, I think -- Yes. I think into compression in AI, it is -- this is a very clear application of AI technologies. It's a very clear way of benefiting and especially in much high large language models were able to succeed based on the large amount of data that is available. Compression also gets better when you have larger amounts of data to model. So I think from that perspective, compression really is right for the picking right now. As with most AI at this present point in time, there's a question of hardware that will allow the end users to be able to use this on their devices. Much like what James was talking about in his presentation is like the hardware is getting constrained. However, there are solutions in the pipeline that will be becoming more and more prominent over the next year -- over the next couple of years? I mean Apple devices already have these NPUs that are pretty much specialized for these neural network model execution. So it's a rapidly growing ecosystem. And there's a lot of innovation happening to hardware space right now as well.

Yes. Just -- that's an interesting one. And then it's just slightly -- just reading the question. There's a few around quantum computing, maybe it's one for you, Mark. I mean where do you think it's too much of a broad question. But where do you think quantum computing fits into this I mean it's part of the puzzle no doubt. But I mean do you have any view on quantum computing in AI or quantum computing is the future of landscape? Well, where does it fit in you think in particular?

Yes. I mean I think we've made some really interesting breakthroughs in quantum computing over the last several years and including companies in the IP Group portfolio. And so we're getting closer to having a useful quantum computer, but we're still some distance from that point and quantum computers will solve a subset of problems or subset of specialist problems. And so I think they will perform part of our kind of future compute in 10 or 20 years' time, but it will be horses for courses and there will be other computing technologies that will we'll sit alongside that, do think it would be great for James to comment on that and how on side.

Yes. Somebody's asked specifically about optical neural link with quantum. Is that a [ proper question ] to make?

Yes. So kind of to be completely clear and clarify, Lumai is not quantum, right? So what Lumai is doing with our optical systems is not quantum in any sense. What the kind of the overlap is, is that quantum will require new forms of hardware that digital electronics doesn't do, right? So up there are many optical computers -- optical quantum computers out there that uses light and uses photons as the quantum domain. You have trapped ions, you have superconducting rings. There are kind of many different hardware implementations of quantum, of which optics this one. But optics itself gives so many benefits, as I mentioned in the kind of the communications industry is the best example of that in terms of energy efficiency and throughput. They're just using optics for classical computing for kind of for the arithmetic needed in AI, you still get huge benefits. So it's very much kind of a classical computing for classical AI as opposed to quantum computing for kind of new quantum algorithms and those sorts of ideas. But I can really agree the quantum will definitely play a role in that's kind of a subset of many different things. You'll have AI, you'll have supercomputer, you have ultra-high precision HPC and you'll have quantum, all kind of working together definitely.

Seb, do you have any comments on that in particular? Does it mean maybe that's a bit far off to apply compression technologies to quantum, but I don't know if there's a link or anything to comment on?

So I think you're very excited about any potential gain potential opportunities that could arise from quantum computing. Right now, you're not really focused on that, that's fear. I think we are -- it's going to be a while before end users are going to have quantum computers. So when that happens, you will lock, unlock a lot of different possibilities. But at this point in time, we are mostly interested in productionizing and really getting to the like getting our methods to work on existing infrastructure.

An interesting question coming up, I think, more towards you said about if the compression delivers a human-centric visual image, how does this compromise AI algorithms that align from a bigger data set improve recognition or bring insights that human observers may find more difficult? So I guess the question is around the narrowing of the data when you compress it and then the input that might be to an AI model. I don't know if you have any comments on that around your methodology and whether that's true or not?

Yes, this is a very interesting question. I don't think we actually know the answer to that. So the point about compression is that depending on what your target application is, you can always tailor the quality. You can definitely change them all based on what makes sense for that particular application. So not all compression algorithms would be great tailored to one specific mode of data. For instance, for this webinar, it makes a lot of sense for really focus on our faces because have the blur background here. We don't real care what it's doing general speaking background are not really that important. And what's more important is for us to be able to be detect facial features and being able to detect micro expression. So that's the key to the quality. And that's also a property of our visual system. But for other applications of data, you might want a different set of features. So if you want to generate data for machine processing, you can tailor your algorithm to maintain a different set of features and they cut different trade-offs in compression. So it's a definitely important question because when I really start messing with the data distribution, you are messing with the future generations of your models, but we don't see this to be a big issue at this point.

Okay. That's really insightful. Thank you. Interesting question. There's a comment around analog computing and somebody mentioning they came across it in 1972, and it's great to see a return. I mean the history repeating itself a bit here, James you're nodding, is it, it is the right time? Is it for analog computing to making a comeback?

Yes. I think the kind of like I mentioned in the talk, the people were very excited to move away from analog because it has these issues in potentially being general purpose computing. It just doesn't make sense, right? You run into issues of how do you do logic, how do you do control all of these things. And the trend was always kind of more and more precise arithmetic, more precise modeling, more precision. And that's what digital unlocks, right? If you can turn any number into a longer string of 0s and 1s as you like. You can effectively have unlimited precision. The cost of that is that you need longer and longer and longer strings are 0s and 1s, it takes more and more time to process. It takes more energy to process. And what we're seeing with AI is kind of the complete reversal of that. People are realizing for AI, you don't need that length of precision. You don't need 64-bit, 32-bit. We're down now to basically 8-bit precision being used in AI and AI models. And if you don't need that level of precision in that really precise kind of flexible control of doing whatever you need. If you only need low precision arithmetic, it doesn't make sense to use that technology, that platform that's really slow and energy.

And just the same argument applied to quantum. That's just a question that's come in quantum systems. Does it -- they don't need perfect precision perhaps. And so this might help air their, this methodology might help aid their implementation more broadly.

Yes. I mean, it's interesting for kind of the interplay between Quantum and AI is really interesting. There's kind of the 2 way around, right? There's AI for quantum. Applying AI ideas to make better quantum machines. Also come we use quantum computing and quantum machines for AI. And those are 2 kind of very different -- 2 very different areas that require different thought processes, but kind of the more exciting one of using Quantum machines for AI is still really new. That's a very, very young area of research because it's not clear kind of the advantages you get from Quantum machines is not clear how to directly map that onto the ideas of AI. So it's a tricky field and a tricky comparison. But broadly, yes, I would say the low noise or the -- sorry, the low precision or the kind of noisy nature of quantum would lean itself well to AI more than 2 other problems.

Great. Any other comments, Seb? Do you have any comment on that Seb or?

Yes. No, I think James started off very well here. I think, yes, quantum is -- being able to use quantum computing in the AI, it seems quite far away at this point, at least in any general capacity. It's a very exciting area of research, but in the future.

Is there anything you came across, Mark? Is that an area you've touched on yet or seen any?

No, not really. No, I agree. It still feels some way off.

Maybe that's a next thing to look out for. Some more slightly easier question perhaps. Who do you think are the big winners and losers in the AI transition? I know that asked a lot. Who are the winners and losers as we move towards more AI and go on Mark, I'll give you the first crack at that one.

Thank you. I'll take the low-hanging fruit and then the others can -- I mean there's a pretty accepted through the big tech is only a relatively small number who can sort of build these big large language models, and they're going to kind of own that space to a large degree, and that's sort of 5 or 6 companies. Clearly, the hardware makers are making [ high ] NVIDIA being a standout example of that right now, but that's just the tip of the iceberg, and I think the implication for James and his for Lumai that does play a key role now. We are not hardware agnostic in this new generation of use cases, and it's going to be a really key element. And then the losers, I mean, the people who sort of fail to invent people who are falling behind because they're trying to do by human power, what can be done by artificial intelligence and all adapting quick enough to the efficiencies of that.

Seb, do you have any comments on that?

Yes. I think the key, the key advantage in the AI world is having the first mover advantage, being able to innovate rapidly and being able to deliver the products quickly. So various large language models. The key success on large language models was placing a lot of research behind a friendly user interface and pushing it all to people that we're able to interact with. . So basically what I would add is it's also important for the consumer hardware to have the relevant accelerators because all of these -- all these models as James alluded to before, we are very, very big and very hard to run in their native form on the consumer device. But eventually, you want to have your personal assistant on your phone. You want to have video compression running on your phone. So we see that when our companies that have thought about placing chips in consumer devices are definitely an advantage in this AI can atmosphere there we're in right now?

James, any views on the winners and loses in the AI race?

So just kind of a similar trend of like this idea of kind of democratization of the compute is really important. If you get like -- I mentioned if you can only run these things in a warehouse filled with 10,000 GPUs and it costs GBP 1 million to build. Obviously, the winners are just going to be the 5 companies even list on one hand if you need this kind of access to the compute in order for everyone to win, right? And that will only come from making the individual the single component quicker, the individual processor quicker. And you can run ChatGPT kind of on a single GPU. Overall, we go out and buy a GPU and running on the computer at home is what you come with, right? So it's not just about filling warehouses with more and more GPU. It's about making the GPU itself quicker. And therefore, what it look like a GPU, it will look like whatever the next generation of hardware looks like be that optical or something else. So -- yes, that's it.

No, that's good it's a good insight. I like the idea of that. One last question before we call it a day. When not people are going to be asking this, I think through and through, where next? What's the next ChatGPT? What's the next, when next generative AI, next field, next application where should we be looking? What's exciting I'm going to see you, Mark, first of all?

Somebody else go first on that.

James?

So for me, it's not -- I think less people say this is multimedia is just ChatGPT is just text right. You put text and you get text out, but it's going to be applying that to images and to video as well. As Seb said, everything is now so rely on video. That's the next big thing is this generative -- video generative imaging for sure.

Seb any other comment.

From the pioneer perspective, I'd say a compression of course. But yes, I second that multi-model, like in terms of capturing the imagination, I think being able to combine all your video and text is the next big challenge and companies are already working on that. So in terms of capturing the imagination being odd as people were ChatGPT, I think multi-model is going to be very interested.

That's great. Thank you. Mark, any final thoughts on that?

Yes. I'm really excited about the sort of the spatial computing, the Apple Vision Pro device at this year, I think, is the beginning of a different mode of interaction with comes then when you think about the combination of AI with that and the sort of computer-generated content that comes with that. We start existing in these that weren't invented by a human. They were invented by I think that's an interesting area for.

That's great. Thank you. Well, I'm going to finish that. We're on bang on 10:00. Thank you, everybody. Thank you, Seb, James and Mark. Really appreciate your participation today. We have some great questions and really good insights there. So thanks again, and thank you all for tuning in and listening to webinar. And yes, you can keep up an update on IP Group news and activity on our website, and hopefully, I'll see some of you round. Thank you. .

Mark, Lee, Sebastjan, James, thank you for taking investors today. Can I please ask investors not to close this session as you'll now automatically redirected to provide your feedback in order that the management team can better understand your views and expectations. This may take a few moments to complete and I'm sure will be greatly valued by the company. On behalf of the management team of IP Group Plc, we'd like to thank you for attending today's presentation, and good morning to you all.