My friend Sabine Hassenfelder just made a video that got nearly half a million views in just a couple of days. Her conclusion? Quantum computers are basically only good for doing one thing, breaking codes. Now, Sabine's brilliant, and she's right that the code breaking progress is terrifying. Google just moved up Q Day, the date in which quantum supremacy takes place, to about 2029, less than three years away. And as I've often said, quantum computers seem to be really good at doing one thing in particular, which is to simulate how quantum computers work. But I think Sabine has missed a bigger story, because right now in my lab at UC San Diego, I'm teaching my undergraduates to build quantum computers and then to program them and then eventually to launch them into space and maybe, just maybe, use them for AI in space, perhaps on the moon. Thanks to Artemis too. You'll hear from these brilliant undergraduates later on, and when you do, you'll see that what they're doing has nothing to do with breaking code.

And by the end of this video, you can do it too, for free. Let me give Sabine her due, because the news this week is really extraordinary. Three papers dropped in a single week. First, Google found an algorithm that breaks encryption 20 times faster than anything we've ever had before. That cuts the qubit requirement from 10 million down to roughly half a million. They thought this was so sensitive they wouldn't even publish the algorithm. Instead, they used something called Zero knowledge proof, basically a math way of proving that trust us, bro, without showing you exactly how it does so. Second, a startup called Oratomic says that they can break RSA encryption with just 26,000 qubits in about 10 days using neutral atom arrays, not the superconducting qubits I'm using in my lab, which are the same that Google and IBM are using.

This is a radical speed up and reduction in complexity. It's awful difficult to get our lab equipment down to just a few tens of millikelvin, just a whisper above absolute zero and far colder than even the CMB, which is what I study at a balmy 3 Kelvin. Now, a third paper by another group showed that they can do it with 10 times fewer qubits than the original estimates required. Sabine is right. This is real and it's accelerating faster than anyone predicted. The researchers themselves are debating whether it's even responsible to publish this stuff. Scott Aronson, one of the top computer scientists alive, said that said, people in the field are reaching the point of wondering, should we publish this or not. In 1982, when I was a wee lad before high school, even accessing a university timeshare computer meant dialing in, often using a clunky acoustic coupler modem.

That transmitted data at a screaming 300 to 1200 bits or baud. The procedure was tedious. Pick up your phone, plug it in, wait for the screeching handshake, type a text based login and issue an arcane command like rmdrc foobar just to navigate a 24 row monitor. That agonizing lag is the perfect analogy for quantum computing five years ago where you waited in a queue for a noisy 2020 qubit result from a remote cloud. Today, my friends at Quantum Rings again, not sponsored allows you to explosively advance on that timeline right now for free. It puts a high fidelity quantum circuit simulator with hundreds of qubits and millions of gate operations right on your laptop, replicating Google's $10 million quantum supremacy experiment on your own hardware. It's really a whole new world and I want my undergraduates and my viewers and listeners in the audience to take advantage of it.

The truth is that quantum computing holds immense promise. I mean unimaginable things. It's very possible that my grandchildren could grow up in a world where cancer is a thing of the past, because quantum computers have provided real time computational simulation to let us experiment with these drugs without the burden of manufacturing them ahead of time, where things like EVs could be four to ten times more efficient, drive as far as you need on a single charge with batteries that were made in a very sustainable way, with materials that were discovered because of quantum computers, where we could optimize supply chains, solving world hunger if the humans can get out of the way. Literally the biggest societal problems that exist today are in reach for quantum computers. And it's not just science fiction anymore. This recent video covers three papers in the course of essentially a week that moved the goalpost dramatically for this goal. And we used to think about this as requiring systems that took millions of qubits, and now we're talking about hundreds of thousands of qubits. And that essentially brings it from like 2035 to 2040 down to kind of like 2029, 2030 for Q Day for when quantum computers will be able to break encryption.

And if it happens in the dark, mysterious things are going to start happening and we won't know for sure that it happened.