Quantinuum Keynote with CEO Dr. Rajeeb Hazra at Quantum World Congress 2024

our next speaker is Dr rajib araj hazra and he's CEO of quanum and has more than three decades of experience in supercomputing and Quantum technical roles all around the world proud of joining quanum he served as the general manager of the compute and networking business unit at Micron Technologies and he spent 25 years at Intel Corporation ation leading the Enterprise and government group technical Computing group the super Computing architecture and planning and Systems Technology research prior to Intel in 1995 he was with Lockheed Corporation and he was based at NASA's Langley Research Center and he Prides himself on building High performing teams with a growth mindset and a culture of Truth and transparency quanum is one of the Platinum sponsors making this event possible and with that would you please welcome Raj to the stage thank [Applause] you good morning good morning morning I hope I don't you understand why I don't stand behind a Podium um it's my first time at the uh Quantum World Congress and it it's great to make some new friends lots of new friends but it's also very good to see some old friends from the HBC and supercomputing communities uh I'll be part of this hybrid journey of quantum Ai and HBC today I'm going to talk a little bit about the main focus and passion that we have at quanum which is bringing the future in uh the road map to a universal fully fault tolerant Quantum Computing nothing could be more exciting like the time we are in in the quantum industry today but we have all different routes and paths to this Holy Grail of being able to serve the use cases waiting for the for the power of quantum Computing useful and usable power uh Quantum Computing I sometimes claim I have the best seat in the house I have the best seat in the house being part of the quantium team we are a leader in Quantum Computing in case for those that are just saying who is quantium uh we've been around for as quanum since uh November of 2021 but our lineage goes way back to about a decade or longer uh as part of honeybell quantum solutions that build the hardware and Cambridge Quantum systems a leader uh in Quantum software and algorithms so bringing the two lead leaders in uh we've got 350 plus of the world's best talent uh scientists and Engineers all the way from Hardware to algorithms uh and we have a growing presence worldwide uh in the US in UK Germany Japan and rapidly expanding to to Quantum hot zones as I call them in places like Singapore I would say we're the Undisputed leader today and I know that a big bold claim to make so um being datadriven and and always providing proof for what we do uh as our culture I'll talk a little bit about what makes us undisputedly leading with a sense a deep sense of responsibility to continue uh to innovate in this industry and take the industry and the ecosystem forward what you'll see on the first row is what we've been doing for the last 10 plus years building the world's highest performing quantum computers demonstrating increasingly improving fidelities up to three n on our commercial systems today and and and Performing record-breaking Feats on industry benchmarks like Quantum volume where we've achieved two to the power 21 on our latest generation H2 56 Cubit machine well that's just Hardware speeds and feeds uh what what are we doing beyond that the second row talks uh about our work to take these kinds of uh physical capabilities and bring resilience and fault tolerance into uh into the fault rapidly improving the usability and the usefulness and the value the utility of these computers with Microsoft we demonstrated recently 800x the largest gap between physical and logical error rates with ACC syndrome uh extraction this is a huge huge feat for the first time and if you think as you should that this was stunning uh wait for the next keynote uh which Microsoft will present themselves we also looked at uh the RCs or the random circuit sampling Benchmark uh and you know this was made a little Popular by Google in 2019 associated with with the word computer Supremacy uh we don't care whether it's Supremacy or not what we did was We performed 100x better on that same Benchmark with our system in doing so we also discovered and and proved that to do the same thing to prove the quantumness of of The Benchmark which is what it essentially does uh we we consumed 30,000 times less energy compared to what a superc Compu would have to be and would have to consume in order to do that same benchmark why do I mention this a lot of people think Quantum Computing is just about doing things faster like an accelerator true in many cases that might be true but it is also true that when we look at the growing needs of AI and the challenges it puts on things like sustainability uh as these data centers start resembling little nuclear power plants um that the the energy benefits the resource efficiency that Quantum Computing and driveing is absolutely absolutely phenomenal at the last row is all about continuing that rate and pace of innovation forward looking at more and more sophisticated era correction in a single shot with 4D codes solving the scaling problems that typically have been touted as what could halt the progress towards fully F to uh Universal Quantum Computing at scale uh with What's called the wiring problem and I'll talk about that later but that's just technology but that's just technology what you'll see on the right hand side is people the leaders in their various n markets whether it's financials whether it's Transportation whether it's life sciences whether it's chemistry all working with us on real use cases and bringing them to life in their next form on Quantum Computing whether we're looking at newy materials whether we're looking at Carbon sequestration with metal organic Frameworks whether they were looking at peptide binding for for new kinds of drugs and Pharma uh Explorations wherever there is a value for Quantum Computing and companies that are at the Forefront of leading that charge because they think it's a strategic Advantage for their future for their business we are there so they these companies are putting their their limited resources in focus and faith in working with us and we are trusted by our strategic investors at the end of the day to accelerate this rate and pace of innovation requires Capital we are very proud that our strategic Partners have conviction in our Direction and technology and they're putting their money to accelerate our Collective Quantum future over the past decade we've had about 1 . 6 billion of capital USD of capital invested to get us to this leadership point and in most recently in January 2024 we completed a 300 million fund raise at a valuation of 5 billion the valuation speaks to our strategic investors belief of not just the opportunity of why Quantum but also why quanum so enough about us what do we do to deliver value what is it we think is necessary in the ecosystem at this point we deliver a full stack we and and the two things that I will point out here is we don't just deliver our stack right we deliver a stack from Hardware to middleware to developer toolkits to algorithms we have experts in chemistry in supply chain in in high energy physics we we take all of that and become a solution development platform for our customers alleviating their complexity of finding multiple partners to fill the stack Gap the two things I will point out is we are very partnership oriented so our software strategy is to be platform inclusive our software runs on more than our Hardware we let our Hardware win on its own merits but we don't write for our hardware and so we have a tool like uh ticket now with two plus million downloads widely being used to develop and optimize code on superconducting cubits um on multiple trapped iron cubits on multiple and and now starting to be on neutral atom cubits as well we also work with others to take the best of their Innovations and make them run best on our hardware and that is again an example around error correction we have our own but if someone comes up with something better or something more tuned to a particular algorithm we have no constraints in our system that says you cannot use that and again if you want more details about this stay tuned for the next keynote so the question is we are building the world's most powerful most useful quantum computer ERS today what's next we believe that our mission our responsibility is to continue not continue to accelerate this track leading with scale with leading scale the lowest error rates and unparallel customization you you've probably heard a lot about the first two let me talk a little bit about the third one we don't think that the world of quantum algorithms as they diversify is going to be simply around a design Point give me these many cubits with this error rate there will be shallow circuits and deep deeper circuits and some of them will do well with full letter correction some won't and so what we are building in our Hardware plus software stack is the ability to customize to an increasingly diverse set of algorithmic quantum algorithmic needs and we think that is important to actually increase the utility of quantum computers and not get them niched to be domain or problem specific it's both a technical point but it's a fundamental economic Point as well because it pertains to the the S you know the basic serviceable available market for a company like us so today I'm introducing Apollo Apollo is both a reality and a vision Ro together in one in this decade we will deliver a quantum computer that supports a universal Gat set with fault tolerant operations that's what we call a fully fault tolerant quantum computer using our qcd trapped iron architecture which has these very critical features like all toall connectivity and midcycle measurement we will deliver thousands of physical cubits at a physical a two Cubit gate fidelity of 10 ^ minus 4 hundreds of logical cubits delivering industry's best error rates and I'll talk about that in a second and a design space where you could trade off between the number of logical cubits and the Fidelity or the error rates you need for the demands of the application dynamically we we think this is a very very important aspect of making a more general purpose quantum computer this will be available no later than 2029 for many this would have felt even a year ago if you go out and read where we kind of get to this this level of performance uh it would be in the 2030s I want you to note the the we can get many people can claim hundreds or even thousands of logical cubits it's really the error rate that then matters right being able to go down and and stay in the 10 the^ minus 5 to going up to 10 the^ minus 10 which we now see through the work of researchers and partners distinctly possible and that is a Tipping Point that is a Tipping Point for the industry that is a Tipping Point for the use of quantum Computing and we believe what makes Quantum Computing come out of the promise and the Hope into a re into a large scale reality for the market Paving the path to AP poloo many people will say and have asked us well do you have all it everything that it takes my answer is yes because in the generation of work we've done we've now demonstrated that we can get to larger uh number of physical cubits by solving the wiring and and sorting problem and we have engineering stats on things like scaling the traps and laser beam delivery it's a little bit more interesting on the fault tolerant scale you can always build machines that have more and more cubits a little bit more easily and particularly working um with making it Universal right so I won't announce it today but I will want you to watch the space of fall tolerant Universal Gat set very clear very carefully we obviously have Clifford gates done we publish that paper watch the space around non- Clifford gates in the next few months and that's all I'll say at this point once we have both of those done we are now well on our way to engineering a polo from here to Polo is not just you know we have H1 and H2 today and then there's a big promise and wait till 2029 what we are doing is we are putting two systems in that are not just development vehicles for a poloo proving out the Trap Technologies proving out the software proving out the eror correction but our also very useful machines in their own right for software development and continue to accelerate the ecosystem around this kind of capability the first system is available it's called Helios will be available in 20 uh 25 June so that's imminently around the corner and we will have 50 logical cubits with an error rate lower than 10 ^ minus 4 so is the next one keeping a 2-year Cadence if you notice everything from H2 is about on a 2-year Cadence um will take that up to uh about 100 logical cubits at an error rate of 10 ^ minus 5 now now it's not just the liit of logical cubits but it's an order of magnitude decrease in the error rate before we get to AP poloo so this represents for the first time our disclosure of a full road map all the way out to 2029 and we are doing this because we are we are looking for Partnerships and the ecosystem to get their parts done and delivered in a way that we can bring a full capability uh to bear in the Apollo time frame The Tipping Point time frame so our job if you go up to our website says it's accelerating Quantum Computing and this era the next three or four years is going to be critical as many of us in the space that we are in will say we are going to do this the next three or four years will prove the habits versus the still aspiring we know on which side we will want to be and we will be so we've actually changed our company Philosophy to include two additional words accelerating Universal fully fault tolerant Quantum Computing that is our journey as a company that is our Collective Journey as an industry and we hope you're both excited and and looking forward to your own part of the journey and working with us to even further accelerate it thank you [Applause]