The biggest challenge thus far has been ultimately error correction. Can you just talk us through what you have just achieved? Yeah. The biggest news for today is that we developed a new, very high-quality quantum chip, and this enabled two

breakthrough computations. And one is error correction, where we showed for the first time that as we use more qubits, the error rate comes down, and it comes down exponentially. The second important achievement, breakthrough achievement, is that we did a benchmark computation that took just 5

minutes, and now a new chip that would take the top supercomputer ten septillion years to perform. So ten septillion—the audience may not know that's 25 zeros—so it's a mind-bogglingly long time, and it

took 5 minutes. What I want to understand is what this benchmark test is—benchmark random circuit sampling. What does this prove? So this benchmark problem—we often say it's not a problem that people on Main Street have yet—but it's a suitable

benchmark to compare different quantum processors to each other or to compare quantum processors, quantum chips, against classical computers. What's so mind-boggling about quantum computing more generally is that we have been talking about it for decades, but it always feels another five or another ten

years off. Is it different this time? Hartmut. Yeah. Quantum computing is definitely not that feared. There's steady progress. And our new chip that we just released, Willow, will form the basis to start doing useful computations that will

enable or solve problems that humankind has. Let's talk about those problems. Then everyone immediately goes to health care, drug discovery. Is that the most obvious place where this sort of computing is going to be helpful? So the well-understood killer

application of quantum computers is to model systems where quantum effects are important, and that is more often the case than you may think. For example, if you want to understand how drugs, pharmaceuticals bind to cells, or if you want to design a nuclear fusion reactor, or something as

mundane as improving the batteries for an electric car, all these engineering challenges involve quantum problems. And so quantum processors are just the tool of choice to solve those problems. But in the perspective of generative AI, the hope that our audience has had over the past couple of years, what

does quantum mean for that? Is it superior even in its effects versus generative AI and some of the same issues in terms of energy and infrastructure? I wouldn't compare it in those terms, as quantum computing will be one of the most transformative technologies of our

time. But there's a lot of cross-fertilization between AI and quantum computing. So AI can help quantum chips to work better and vice versa. There are many key computational tasks in AI where a quantum chip would just be the tool of choice to handle those.

Google Computer Solves Septillion-Year Task in Minutes

Welcome to the Bloomberg Technology YouTube channel. Subscribe for original series and mini-docs on game-changing technologies, profiles on innovators, live conversations with experts, and breaking technology news & analysis from Bloomberg TV.

Google Computer Solves Septillion-Year Task in Minutes

Google Computer Solves Septillion-Year Task in Minutes

Scripts