Google recently announced that its 53-qubit computer performed a task in about 200 seconds that would have taken about 10,000 years for other supercomputers. That’s kind of a big deal! But not everyone agrees – shocker – we have a techno-clash of the computing titans going on. In this two-part episode of The Disruptive Enterprise Podcast, host Greg Turner discusses the potential impact of quantum computing supremacy with guest Bill Kleyman.
Has Quantum Supremacy Finally Been Achieved? (Part 1)
With Gregory J. Turner & Bill Kleyman
Gregory J. Turner: Welcome to our continuing podcast series at The Disruptive Enterprise. Today, we’re going to continue to talk about disruption but, specifically, again, about quantum computing in light of some very new and recent information. And of course, to help me explore this topic is Bill Kleyman. Bill, as a brief reminder, is an advisory board member for MTM Technologies, and he is the Executive Vice President of Digital Solutions at Switch. Welcome back, Bill, to the Disruptive Enterprise.
Bill Kleyman: I am so excited to be here, it feels like it’s been forever since we talked last time, Greg. And I miss this interaction. And all of a sudden, out of nowhere, I get an email saying, “Bill, guess what I want to talk about some more.” And I’m like, “Well, what is it, Greg?” “I want to talk about quantum computing.” So, apparently, this is still an interesting topic for you. And I’m really glad that it is, Greg. So, I’m curious, what’s on your mind when it comes to quantum computing these days?
Gregory J. Turner: Well, Bill, I guess the question that is on a lot of people’s minds but, you know, specifically, my mind is, did Google reach quantum supremacy?
Bill Kleyman: All right. So, we’re going off of—let’s let all the listeners here kind of get kicked in a little bit. So, Google recently announced, I think just like in mid-October, that it came out, and its 53 qubits computer performed, in about 200 seconds, an arcane task that would have taken about 10,000 years for a different supercomputer built by IBM, none other than, roughly, like I said, 10,000 years to actually complete. And it has been ranked one of the world’s fastest and most efficient supercomputers out there.
Bill Kleyman: So, all of a sudden, we’re in 2019, haven’t even hit 2020 yet, and we’ve got somebody on the top of the mountain waving the quantum computing supremacy flag. That’s kind of big news, man. That’s kind of a big deal. Despite what we’re going to be talking about here moving forward, it’s a really amazing scientific achievement, where Google called their shot, they came out on top, they’re saying it’s a big step forward towards actually removing any of these plausible arguments that make quantum computing impossible, and we did it. We’re up here. We did it. It’s amazing. 52 qubit computer, 200 seconds, arcane task, it’s amazing. So, everyone’s out there celebrating and thinking it’s amazing.
Bill Kleyman: Now, do I think that they reached true quantum supremacy where some of their machines were actually doing quantum entanglement, really generating those powerfully random numbers that they’re able to actually do these compute cycles? I’m on the fence. I’m really on the fence. And maybe even leaning towards a little bit saying that I’m not sure if they necessarily did because the follow-up report—and obviously IBM came out saying, “Whoa, whoa, whoa! Pump your quantum brakes here for a second, because I don’t think you did necessarily what you think you did.” And they said—it was just a couple of days ago, right? October 21st, they came out and said that by tweaking their own Summit machine, their own supercomputer, making some changes, they were able to do it faster, about two and a half days effectively, saying that this threshold for quantum supremacy, doing something a classical computer can’t still has not been met just simply by doing some tweaks on the Summit machines.
Bill Kleyman: So, the argument becomes, have we really created a machine that’s capable of maintaining something known as quantum coherence? That actual state where the quantum bits actually operate together. Remember, it’s not just getting one of those qubits to work together. The more you have, the more challenging it is to maintain quantum coherence. I’m still going to go out there and say it’s a very, very elusive concept. Maybe we’ve scratched the surface of it, and we’re seeing what’s truly possible around quantum computing. But again, in reality, quantum computers continuously are beset by errors that always need to be corrected for. And it’s devilishly difficult to maintain these qubits for any length of time. So, they tend to decohere. They lose that delicate quantum nature and that quantum state. Much like smoke rings break up at the slightest air current, that’s what’s happening with these qubits.
Bill Kleyman: So, these guys are waving the flag at about 200 seconds. And I’d love to pop the hood open and see, did you really maintain quantum coherence? What kind of error correction are you using? And it’s also really, really important to note, it’s not necessarily a perfectly functioning quantum computer here, especially if IBM really did come back and say, “We tweaked the supercomputer machine here, and we’re able to do some of this other stuff that you say only a quantum computer can.”
Bill Kleyman: Because we have to remember, everybody listening, I love this concept. I think that we’re getting really sort of excited and ahead of ourselves. Maybe not unlike when we start talking about like 2G, and 3G, and 4G, and 5G kinds of traffic, when everyone’s like “LTE is kind of advanced out there.” Is that 5G? Well, not really. It’s only maybe between 4G and 5G.
Bill Kleyman: But again, these machines themselves, they’re extraordinarily delicate. Really, really delicate. And engineers have to isolate the quantum chip from any sort of noise in electrical, magnetic, thermal, it doesn’t matter. And just the temperature in the room can render the whole machine useless. So, aside from error correction, aside from we can’t be completely sure of every calculation these quantum machines make, you can still—obviously, you have to consume this extraordinary power, this extraordinary power, and you still have some of these mistakes.
Gregory J. Turner: Right.
Bill Kleyman: So, I don’t know. Deep neural networks are something that I’ve been taking a look at quite a bit where they use mathematical complex systems to learn discrete tasks. But this is where these kinds of science tools become a real play. So, Greg, I want to hear your thoughts on this. I’m on the fence. I don’t know if they necessarily reached quantum supremacy. And if it really is true that IBM came out, and they did cast their doubt, and they’re saying, “We were able to do this with a classical computer,” then, I don’t know. I don’t know, man. It’s-
Gregory J. Turner: Sure. So-
Bill Kleyman: It’s one task that they’ve done.
Gregory J. Turner: Yeah. Well, let me ask you this. So, IBM’s position is based on the definitions that people are using for quantum supremacy and quantum computing, they’re saying, “Hey, if we’re able to do this with a traditional computer model, then it can’t really be the example of a quantum computing solution or quantum supremacy.” However, it seems to me that if we’ve taken something that until IBM tweak something, in finger quotes, it would take 10,000 years to do, or 10,000 hours, or whatever the large sum of numbers was, and we’re able to do it in 200 seconds, that seems to me that it’s an order of magnitude improvement in computing. And whether or not Google was able to do it with 53 qubits working with complete coherence, or there were 48 of them, or 49 of them, seems to be immaterial to me. The result seems to be greater than the process by which you got there. What do you think?
Bill Kleyman: So, I like where you’re going with this. You’re suggesting that we have evidence of quantum supremacy and evidence of quantum computing. I’m not denying that. Despite the fact that it’s elusive, it’s certainly possible. And again, maintaining that quantum state, if anybody is listening to this, and you haven’t Googled yet what a supercomputer looks like, again, I’m going to reiterate what we talked about last time. I absolutely think it looks like a chandelier that should be in only Stark’s house. But these things are amazing, and they look really, really cool, but they also operate in extraordinarily conditions.
Bill Kleyman: I’m not sold on this because, again, the impetus for change happens when other things occur in nature in our world. So, you’ve got Google who came out and says they’ve created a quantum computer that’s capable of doing this calculation. Well, IBM came out and said, “Well, the thing is let me take a look at what you were doing here.” And maybe this wasn’t their focus. Maybe that 10,000-year-old equation or process wasn’t something that they were necessarily looking at. And they started to focus on it as soon as Google came out and made their statement.
Bill Kleyman: I want to go back, though. I don’t want to sound like I’m the negative guy here, and I’d only get an e-mail from my friends at Google saying, “Bill, what’s up, man? We thought we were buddies,” because we are. And I’m a big fan. I just want to kind of stay on the centrist path, where I’m a little bit more cautious as far as the approach. I feel like maybe we did reach a milestone. But if this quantum machine was truly operating in its state, quantum supremacy, as they would call it, I’d want to see more. I think I’m the kind of person that says, “You’ve got to win. Good job. We’ve heard a detraction from that win from one of your competitors but win nonetheless.”
Bill Kleyman: It is like I said earlier, it is a great scientific achievement. It’s a big step towards actually creating a machine that could do quantum processing but, again, I want to make sure that everybody does understand that quantum machines, it’s not necessarily designed to do what a classical computer might need to be able to do. These are really advanced mathematical equations. These are really advanced genomic formulas that they might be looking at. So, it’s not just out there to play Halo, or Grand Theft Auto, or whatever you want to do. It’s really designed for very specific types of mathematical computations. And to be quite clear, there are classical computers that can do those mathematical computations better than a quantum computer, even in its quantum state. So, it’s early. It’s early. I’m excited that people are waving the flag. I really do think this is going to be something in the near future.
Bill Kleyman: I mean, listen, back in 2013-2014, I wrote an article around quantum cryptography for a defense magazine. Actually, for the United States Government. It was a defense magazine. And it was the first time ever that they were able to do a line of sight, point-to-point, don’t stand in the middle of it, an actual quantum cryptographic encryption between two points. Obviously, this is an application for the modern network, but it gets really much more complex if you take it like look at a layer, two-layer, three networking. Obviously, you turn a corner, and this thing isn’t going to work anymore. But that was in 2013. We’re still developing elements of quantum cryptography that we can be able to use potentially later on.
Bill Kleyman: It’s a slow process. It really is. It’s something to get excited about. But what I’m—for me, Greg, and everybody listening, I’m going to be sold when we can actually create real practical use cases and results. Like Google comes out in a week and says, “We were able to create a genomic sequence for this healthcare patient that was able to create the most unique cancer drug treatment ever potentially possible, never before being done by a classic computer, nor is it possible to do this with a classical computer.” That is what I’m going to be sold, where we’re finally using use cases or deploying use cases, where there’s like there’s no question to get necessarily. Sure, maybe, there might be some questions, but when we get to a point where these things are actually doing something, I don’t know, Greg, maybe I’m being a little bit too a pessimist, which is really not like me at all.
Gregory J. Turner: No, no. I don’t think you’re being pessimistic. I think that’s—with anything that’s being developed, I think we need to look at the benefit of what it is that’s being developed. And I think picking a really good use case like some kind of a cure for a disease or some kind of mapping of the gene sequence that will help us in developing a cure, I think, the right kinds of things and very noble. I think from my standpoint, I’m going back more to my theoretical mathematical days in college and computer science days, and I’m kind of hung up on the notion of whether or not a classical computer could do it or not.
Gregory J. Turner: And I guess from my standpoint, have we thought of every condition? As humans, have we thought of every condition known to computer science of what a classical computer could do or not do? And certainly, aren’t those conditions really the basis of how we do investment in technology today? Some of our partners, like Cisco and others, have created an incredible parallel processing and core processing to help with network management and network security. And those advances, 5 years before or 10 years before, would have been thought to be impossible in the computer framework because of the way it manages heat inside the device that I think I feel like it’s too limiting of a definition, Bill. And that’s where I’m kind of looking at this.
Gregory J. Turner: I’m excited that a major player, like Google certainly is, has made investments, has taken the risk to create a framework where quantum computing can be researched, and developed, and continue to help open some doors and create new ideas that could introduce more development, so that we go from 53 to maybe 124 four qubits, and so forth.
We are going to take a break here. Please join us for Part 2 (Episode 6) of our conversation on Quantum Computing.
MTM Technologies, Inc. is a Citrix Platinum Solutions Partner. You want leading solutions from Citrix – XenApp, XenDesktop, NetScaler, and more. Let us help you architect, build, deliver, train, and support them! We are the largest North American Citrix Solution Advisor, holding all major Citrix certifications and earning numerous awards.