The Vergecast - Time to believe the quantum computing hype?

Episode Date: July 9, 2026

Quantum computing hype is everywhere. Trump says we'll have one by 2028, Microsoft by 2029. IBM is investing billions. Is it really happening? Science journalist Sophia Chen joins us to discuss her article, "What is a quantum computer good for? Absolutely nothing — yet," and break down what's real and what's hype in the quantum computing race. Further reading ⁠Meta is reportedly working on smart glasses that would be recording all the time⁠ ⁠Sonos laid off some of its veteran product and design executives⁠ ⁠Cash App fraud⁠ ⁠Starlink deployments on record pace⁠ ⁠Play-Doh for adults is the new Lego for adults⁠ ⁠Slate is hoping a little Razzmatazz will help sell its trucks⁠ ⁠What is a quantum computer good for? Absolutely nothing — yet⁠ ⁠A new paper argues Microsoft exaggerated its quantum claims a year ago⁠ ⁠Drama over quantum computing’s future heats up⁠ ⁠The race is on for quantum-safe cryptography⁠ Subscribe to The Verge for unlimited access to theverge.com, subscriber-exclusive newsletters, and our ad-free podcast feed. We love hearing from you! Email your questions and thoughts to [e-mail] or call us at 866-VERGE11. Learn more about your ad choices. Visit podcastchoices.com/adchoices

Hello and welcome to the Vergecast, the flagship podcast of Probabilistic Computing. I'm Jay Kastranakis, Executive Editor of the Verge, and today we are talking about quantum computers. Quantum is hot right now. Trump just issued an executive order saying that America, quote, stands at the cusp of a quantum revolution. The order demands that America accelerate deployment and commercialization of quantum computing. Meanwhile, his science advisor promised that the U.S. government will develop a computer, quote, powerful enough for scientific discovery. by 2028.

Meanwhile, Microsoft just unveiled Myerana 2, its second-generation quantum computing chip. Last year, Microsoft CEO, Satcha Nadella, said that its team had created a new state of matter that had only been theorized 100 years ago. This year, the new Myerana 2 chip has apparently, quote,

put the team on a path to achieve a scalable quantum computer that is commercially valuable by 2029. These are big, ambitious goals. Like I said, quantum is hot right now. There are a lot of big promises and a lot of money flowing into the space. We were really interested in what's real and what's the hype. So we had science writer Sophia Chen dig into this for us. She's joining me today along with Verge tech editor Marina Galaparina.

But first, here's what's happening on the verge today. This is 90 seconds on the Verge for Thursday, July 9th, 2026. It's only July, but Anthropic has launched with my colleague Hayden Field is calling Claude Rapp. It's a new dashboard that shows how much time you're using. using Claude, your top days, how much time you're spending, that sort of thing. If you're using Claude enough to bother checking this feature, you might want to take a break. Fortunately, Anthropic is also adding a feature that can send you reminders to sign off. So maybe take a look at that. Meanwhile, OpenAI is finally upgrading ChatGPT's voice mode with a new model that it says is far more

capable. I don't want the needles to be too big. What do you think? If you go up, it's going to get kind of baggy. Although baggy is very popular now, don't you think? Chat Chatsy-GBT's voice mode has been on a two-year-old model until now, and it's been pretty bad. You might have seen that one guy on TikTok who gets ChatGBT-Btie to give dumb or bizarre responses. The old model is part of the reason why.

I think that guy might have single-handedly embarrassed Open AI enough into upgrading it. Open-Ai even gave him early access to test the new model, so we'll see if he can break it. Finally, Character AI, the AI Chatbot role-playing service, is getting into AI-generated vertical video microdramas. I watched a couple episodes. It only took a couple minutes, you know, and they completely break my brain. I think everyone should watch these just to understand how bewildering they are,

and to recognize what the end state is of TikTokified attention spans. You can read more at Theverge.com. That's 90 seconds on The Verge for Thursday, July 9th, 2026. All right, this seems to happen every few years. Noise around quantum computing has reached a fever pitch. Everybody is super excited. There are press releases left and right.

about new quantum computing developments, and then nothing happens most of the time. So just to start us off, Sophia, why is there so much attention on quantum right now? I think that a big part of what's driving the interest is there's this Cold War-esque framing where it's China versus the U.S. over quantum computing. And then in addition to that,

the big tech companies are investing billions of dollars into this new technology. You've got Google, IBM, you've got these smaller startups, you've got Amazon, all these big names, and then periodically they'll come out with like this real flashy press release about what they've done,

and like 80% of it is like gobbledygook, but it seems like everybody is really excited. This is a question that we get from commenters all the time who are complaining they don't understand what we're writing about. What is a quantum computer? How is it different than a classical computer. A quantum computer is a fundamentally new type of computer, and so they

already exist in an experimental form. So Google, IBM, Amazon, and academic physicists have already made quantum computers, but they're these very small-scale prototype devices. And they're not meant to be consumer gadgets, and it's very likely that actually a quantum computer is going to be this specialized data center that you log into via the cloud. What makes a quantum computer? What makes a quantum computer, computer different from a classical computer. So by classical computing, I mean like your laptop, what's in your phone, like GPUs that train AI. These are all classical computers. And so for a classical computer, the basic mathematical language is binary. So ones and zeros. And so to run any piece of software, your classical computer has to translate the commands into ones and zeros. The computer

represents the ones and zeros using transistors, where a one corresponds to a transistor that's or zero corresponds to a transistor that's off. A quantum computer, on the other hand, doesn't use binary bits at all. It uses quantum bits or qubits, which are probabilities of zero and one. And as an analogy, I like to tell people to imagine flipping a coin.

Before the coin lands, it's neither heads nor tails. It's a probability of both. It's in this in-between state. So a qubit is also in this in-between state, like the coin flipping, it's neither a one or a zero, but a probability of both. And so quantum mechanical objects like electrons and atoms, they exist in these probabilistic states, which is why we call this a quantum computer when it deals with probabilities like this.

It's a computer that obeys the math of quantum mechanics. And when you represent information this way, you can do different types of math that are difficult for a classical computer. This is what I think is so fascinating about these. I think our current model for computing is so based around consumer electronics, right? Obviously, these high-powered data centers exist today, but ultimately it's a lot of the same technology just scaled up. And this idea of having a specialized computer that operates on its own is sort of a different model, right? but these things, they operate at, like, deep, deep, like close to absolute zero temperatures, is my understanding.

And I think we are entering a situation where we only need a few of these to actually work to get the benefits. Is that right? And then the question is, what are those? What are we trying to achieve here? Okay, so actually, that's a common misconception about the cryogenic temperatures.

So some quantum computing designs do require absolute zero temperatures, but then it really depends on the qubit hardware. So the stuff that you're thinking of is made by IBM and Google. They're these superconducting qubits and the picture of that chandelier. It's those things that need to go in the cryogenic temperatures. Whereas there's other emerging platforms, so like using atoms, so like those have different requirements. And my understanding is that those do not need cryogenic temperatures.

That's an argument for the company. that are doing that, you know, that'll be easier to scale. And actually, Google has recently made an announcement that they were going to try doing neutral atoms in addition to superconducting qubits. Definitely, like, in terms of the metrics, so like, you know, when we talk about silicon chips and transistors, we're talking about like 100 billion transistors in the size of a fingernail. Like, it's like a totally different scale. So like when we're talking about qubits, like, we don't need that many, but actually one of the big questions is how many do we actually need to make something useful? So right now, companies are making things in the hundreds of physical qubits.

Cubit is both the unit of information and also the name of the hardware. And what they're finding is actually a unit of information needs multiple cubits because that's their way of reducing errors. And so, yeah, that's something that people are working on figuring out right now. And they're also trying to figure out, you know, what applications are good for the computers that they currently have. So you've interviewed a lot of physicists for this story. Have they told you if quantum computers are good for anything yet? No. They do not think they're good for anything. The current, the existing quantum computers are not good for anything yet. Right. So who's invested in this? hoping to benefit? So people are so excited about quantum computing because it's a completely different

paradigm for computing, and they've come up with all these algorithms, which are simply too difficult to implement now, but they are very exciting. I guess the most near-term application, I would say, is in molecular simulation. So using the quantum computers to simulate complicated molecules and to simulate chemical reactions. And these are really interesting to the pharmaceutical industry for developing new drugs. It's also interesting for maybe EV companies, like people who are making batteries, like different material science applications. And so that's one goal application.

And then another goal application is they seem to be good at optimization, potentially. And so, you know, that could be useful in any sort of industry that, you know, does logistics. And then also the banks are interested in it for financial forecasting. But how close are we to that actually happening? So, okay, so we're, that's a difficult question. It's, we're probably a while from a lot of these applications. The way that I like to think about it is like these physicists have written like a very beautiful symphony and they just have, they just have like a couple crappy clarinets. Like, they can't play what the beautiful music that they've prepared. They have to improve the hardware. And I want to

give them some credit because it's actually really, really difficult. It's quite different from like working in classical computing because they have to build everything from the ground up. They have like basic material science problems. They have to solve, like how to make the chips, like what combination of metals and semiconductors to use for different parts. They have to use customized lasers. Like, there's very little, like, off-the-shelf equipment that they can use to make quantum computers. Like, and you mentioned cryogenic cooling earlier. Like, some of them have to have these special refrigerators.

And they also have to figure out lots of basic engineering questions. Like, how do we want to arrange the wires so that they don't interfere with each other? like very basic things like this. And so it's really, really difficult, and they are making progress. But a lot of these applications, they just aren't going to happen anytime soon. The thing that I think is so hard to untangle is the difference between the hype and excitement that these companies come out with, whenever they have any little advancement, which, in fairness, I'm sure these are very, very hard problems.

And it is very exciting that they're trying to tackle them. But, you know, this is like two years in a row note. We've had Microsoft come out and be like, we have this new chip. It's exciting. It's getting us so close to quantum. And then when you actually dig into what they've done, you read these quotes where it's always revealing just how far they really are. So with the Myerana 1, there was this line where they said,

this architecture offers a clear path to fit a million cubits on a single chip. They're currently at, I think it's eight cubits. right? So they have a path. Myerana 2, right? Once again, the team is on a path to achieve a scalable quantum computer. So the Myerana one was eight cubits, and they're trying to get to one million. And I guess my question is, like, how do we weigh these two different statements, right, where they're putting out these very aggressive dates, right?

Trump wants a quantum computer by 2028. Microsoft wants something by 2029, right? Is a scalable quantum computer actually just some technology that they say can scale by 2029, or is it a functional thing that can be used by 2029? And I think this is where it gets hard for the layperson to understand what is happening here. I think in this conversation, it's important to say that when it comes to overhype, Microsoft is in a league of its own. So basically, like, Google and IBM, like, I mean, I would definitely say like they also have hyped

things, but like the thing that people complain about Microsoft about is like whether or not their quantum computer even exists. Whereas for Google and IBM, people don't dispute that. Yeah. Great news. The federal EV rebate is back. Eligible customers get up to $5,000 with the federal EVAP rebate on select 2027 volt and 2026 Equinox EV models. Visit your local Chevrolet dealer today for more details. Before the next track starts, have you ever wondered, who are the people in my old family photos? Or what brought them to Canada? With Ancestry, you can start finding answers. Start with a name and we'll guide you from there. For just

499, try our new 30-day Beginners Pass and uncover generations of family stories. No experience necessary. You've got questions? We've got Ancestors. Offer ends July 22nd, 2026. Visit Ancestry.ca slash beginners for details. Terms apply. I'm Seth Matlins. My new show, Creator Destroy Reimagining Marketing, explores how every decision a company makes, not just the marketing ones, but the HR, IR, pricing, org design, and planning ones, the ones most don't consider marketing at all, contribute to either creating value or destroying it. Each week, I sit down with CMOs, CEOs, founders, cultural thinkers, the people building, breaking, and reimagining how businesses grow or don't for conversations about what

creates value and what destroys it. It's a business show. It's a marketing show. Creator destroys the show that argues. They've always been the same thing from the Vox Media Podcast Network and the Wisdom is company. New episodes drop weekly on YouTube and your favorite podcast app. I think it's less a question and more, you know, how do we, how are we supposed to understand this distinction between the hype and the real achievements they're making and the the realities of the fact that they have eight qubits and they want to reach a million in like two or three years? I guess in the case of, I'll just use IBM as an example because I think they are more of a realist. They have come up with a roadmap where they plan to make this data center by 2029.

And it's unclear whether or not this size of computer is going to be useful. One of the big problems previously was that the more cubits you attach to each other and use for computation, the more errors there were. And so they've figured out that problem at a small scale where they're like, okay, we can build a more complex quantum computer and not have it create more errors. Instead, there are fewer errors the larger we make it. So they have like a roadmap where they talk about this. I think for the layperson, one important thing to keep in mind is just like while they say they're scaling this up, to always ask like, what are the applications at that stage? And so for IBM, by 2029, they want to make this data center and they want it to be made of 200 logical qubits. So basically, it can encode 200 units of quantum information.

When I talked to other experts, they were like, we're not sure what that would be good for. But at the same time, there's these algorithm developers who are also trying to figure out, like, is there a simpler algorithm that might be interesting for industry? And so, like, there's a lot of development all happening at once. I think one of the biggest misconceptions is that a quantum computer is a classical computer that is faster. That is just not the case. And it's not going to be good at a lot of things classical computers are good at. Like I mentioned, like it's not going to be good at email. It's not going to be good at word processing.

It's not going to be good at, you know, this podcast recording, like, live streaming. It's not good for any of that stuff. It's supposed to be good at molecular simulation, which supercomputers, like, require a lot of compute to do right now. And these simulations take a really, really long time or they're prohibitive. And so, yeah, so it's a completely different class of computer. The one thing we haven't gotten into yet is encryption. And it's funny because that feels very different to me from molecular biology and, like, interesting and important science research. But I get the impression.

And Sophia, I'm curious from your perspective. But my impression is that the encryption issue is a lot of why there is this arms race around quantum right now. The encryption issue is very convoluted to talk about because basically it was an argument to make quantum computers, but also it was also a reason to be scared of quantum computers as well. So in 1994, this computer scientist, he developed a quantum computing algorithm. This was before any quantum computers even existed. So his name is Peter Shore. He came up with this algorithm for factoring prime numbers. And so the current encryption system that we use, the RSA family of algorithms, they rely on computers being really, really bad at factoring prime numbers. And so the fear was that once quantum computers existed, that they would be able to break all RSA encryption. And currently, none of the existing quantum computers can do this.

Basically, that was a real motivator for cryptographers to develop better encryption. We have what's known as post-quantum cryptography. So these are algorithms, encryption algorithms that quantum computers are not supposed to be able to break. And they're not in widespread use yet. One of the executive orders that Trump signed recently, he was ordering that the government computing systems migrate to post-quantum cryptography by 2030 or 2031. Is that going to be necessary? Do you really think that that's like a possibility? I do think that it's probably a win for encryption.

I don't know what it means for quantum computers, but yeah, I think that, yeah, I think it's probably a good thing for encryption. Of course, the question everyone's been wondering about, is U.S. going to be China? Well, China is doing quite well. The guy to watch in China is this guy named Pan Jian Wei, and they are using photons to make a quantum computer, And they also had this quantum satellite that they were doing things relating to a related technology, quantum sensing and also quantum cryptography. And, I mean, I don't know. I don't know. I don't know what it means to win.

I mean, it is this interesting thing, right, where I think because this hardware is so specialized, we don't need hundreds or thousands of quantum computers. we need a handful of capable ones. And if you can crack encryption, then that's great. The government probably just needs one of those computers. Getting there first seems really important, if that's the case. And it's so interesting that the U.S. government is putting out these, I would say, outrageously ambitious timelines to get to a quantum computer. It's still, from the sound of it, feels like we are at this R&D phase,

where perhaps there actually has been some meaningful movements. over the past few years. But it seems like there is still a large gap before the stuff is actually like implementable and usable in a meaningful way. Well, there's a confusion because the flashy announcements don't cover the really incremental esoteric things

that Sofia addressed in the article because they're not as interesting to the layperson because they're confusing. Well, should we talk about that about the fact that there's at least one physics researcher who Oh, the drama. Right. They think that Microsoft

maybe didn't even create what they said they created. Right. It's the same thing that happened last year. The girlies are fighting again. Yeah, Sophia. Is that something that we should you know, take seriously? Yeah, I think so.

So basically the critique was that, yeah, Microsoft did not create a myer on a particle, which is the basic building block of their quantum computer. In June, Henry Legg, this academic physicist, he published a peer-reviewed critique. And it was a criticism of Myerana 1 from last year, but he says that the same critique applies to Myerana 2.

And he's not the only one. I've talked to other physicists as well, who have a lot of criticisms about Microsoft's approach, and in particular that what they write in their papers does not match their announcement or their plans for scaling. So that's the Microsoft situation. Google and IBM, you said, are a little bit more sober in their approach. What are we expecting from them in the next couple of years?

So for the last couple of years, Google has been doing these what they call quantum advantage experiments where they're doing these algorithms which don't have much practical use or any practical use, but they are trying to show that they can do it better than a supercomputer. And at the same time, this year they also announced that they are going to make a neutral atom quantum computer. So their current quantum computing chip, Willow, is made from superconducting circuits. And so now they're branching out to a different hardware platform. With IBM, they are planning on making a data center sized quantum computer by 2029 with 200 logical qubits. And I think at the same time, everybody's just, they're working on improving error correction. So this idea that, you know,

they need a special technique to correct errors in a quantum computer. And so being able to use fewer cubits to represent one piece of information and then also figuring out how to do that at scale. So yeah, I think in the rest of the industry, they are working on figuring out how to do error correction better and make things bigger. You know, and I should say Microsoft's, of course, disagrees with those researchers' criticisms of their work. I think what's really interesting right now is for so long, I think this stuff has been really hyped as just around the corner. and we are actually seeing hard dates this time. And we will kind of just be able to answer those questions

in a couple years' time, right? Like, I think before it was always, this stuff is right around the corner, but when is around the corner? And it sounds like there is progress happening. I'm not sure if I believe how aggressive they are. Well, they're using terms like practical and scalable, both of which have very specific meetings, right, Sophia?

I don't know that I think it's way well I guess like by practical I in my definition it means something that they wouldn't be able to do with a non-quantum computer that also has some sort of commercial value and by scalable it means I think by scalable what they mean is that they can build them bigger without running into this. problem of more errors. There are some bullish researchers, like these are, someone I spoke to in academia who was like, I think that we could do a scientific simulation by 28 that was, you know, scientifically interesting, might not be a commercially interesting simulation. And so she was studying like a simplistic model of photons interacting with electrons, which applies to photosynthesis. And it also applies to solar cells. And so she was very optimistic. I talked to some other people who were like maybe by 2030, 2035.

And then I had another guy who was just like, I feel that they have totally underestimated how difficult it is to scale. Like I think it's going to be a couple decades. So the responses to that are all over the map. So you think we're going to write the same article next year and the year after? Yes. And actually, I think it'll be very similar. And I think the only thing that will change is that there will be like one little section in there about the nerdy incremental stuff.

Because you probably won't let me write more than that because it's pretty boring. Sophia, I really do want to have you back every single year to write this. So did they do it again? And I know it's going to be the same for a while. But I find this stuff really, really fascinating. And I think the upside is like this stuff is exciting. exciting. As they get closer, I think the possibility of even these, you know, quote-unquote, scientifically interesting experiments, that's very meaningful, and that is a start. But cutting through,

I think, their very excitable language about how close they've gotten is very difficult, particularly with the subject, this naughty and dense. So I appreciate you coming on and speaking with us. The story is fantastic. If you all haven't read it yet, check it out. It's on the birch.com. I'd want to say thanks to Sophia and Marina for joining me today and thank you all for listening. If you like what we do here and want ad-free podcasts, you can become a paid subscriber to The Verge at theverge.com slash subscribe. The Vergecast is produced by Josh Kajas, Eric Gomez, Brandon Kiefer, Travis Larchalk, and Aaron Locasio. David is back tomorrow, or is he? See you later.