New quantum arrangement Google Willow - a revolution, but with a catch? (film, 5m)
The video on Fireship discusses groundbreaking events in quantum computing, focusing primarily on the El Capitan supercomputer and Google's new quantum chip, Willow. El Capitan, the world's largest supercomputer, operates on traditional principles, boasting millions of CPU and GPU cores, while quantum technology is rapidly becoming more dominant. The Willow chip, though small enough to fit in the palm of one’s hand, can solve certain problems at a speed that vastly surpasses that of El Capitan. To illustrate, a problem that takes Willow 5 minutes to solve would take El Capitan an unimaginable number of years, greater than the age of the universe. Such comparisons are both astonishing and fascinating, but they also underscore how far we are from practically deploying this technology.
Fireship points out that the United States isn't the only player in the quantum computing space. Recently, China unveiled a record-breaking 504-qubit chip, sparking significant interest online. Many began to realize the potential threats such powerful computers pose to data security, as they could crack encryption algorithms that were previously thought to be unbreakable. This prompts the video to analyze these dangers while also explaining why, despite significant advancements, quantum technology is still far from being genuinely useful.
The video introduces the principles behind quantum computers, which differ fundamentally from classical ones that rely on bits. Quantum computers utilize qubits, which can represent both 0 and 1 simultaneously, enabling them to perform calculations in parallel. Fireship explains that qubits can become entangled, which is yet another strange aspect of this technology. Despite its complexity, qubits are incredibly sensitive to disturbances, leading to frequent errors in computations. This poses a considerable challenge for the development of quantum computers.
The video explores some of Willow's achievements, including its ability to improve the accuracy of qubits and extend their superposition state duration to 100 microseconds. Although this chip has only 105 qubits, it can already execute certain calculations significantly faster than classical computers. It also highlights that those accustomed to traditional computing should still await a breakthrough in quantum computing that is likely when the number of qubits grows to around 2000.
In conclusion, the video emphasizes the importance of a low error rate in qubits, which will enable the growth of quantum technology. At the time of writing this article, Fireship's video has garnered over 1.6 million views and 59,968 likes. This publication attracts attention from the tech community and encourages further discussion, especially regarding the future applications of quantum computers and the potential threats associated with them.
Toggle timeline summary
-
Introduction to El Capitan, the world's largest supercomputer.
-
Google reveals its new quantum chip, Willow, that outperforms El Capitan.
-
Willow can compute certain problems vastly faster than any classical computer.
-
China also announces a 504-qubit superconducting chip.
-
Discussion on the implications of quantum computing.
-
Quantum computing's potential benefits and threats to humanity.
-
Introduction to the concept of how quantum computing works.
-
Comparison between classical computers and quantum computers.
-
Explanation of qubits and their ability to represent multiple states.
-
Entanglement of qubits and its significance in quantum computation.
-
Challenges faced by quantum computers, including error rates.
-
Google's Willow chip features innovations in qubit error management.
-
Potential of quantum computers to solve problems like prime factorization.
-
The necessity of reaching a specific qubit count to break encryption systems.
-
Concerns over error rates in emerging quantum technologies.
-
The timeline of advancements in quantum computing since 2016.
-
Promotion of the daily.dev platform for developers.
-
Closing remarks and invitation to join the developer community.
Transcription
This is the world's largest supercomputer, El Capitan, with over 1 million CPU cores and 10 million GPU cores at its disposal. Some might call it blazingly fast, but Google just unveiled a new quantum chip named Willow that can fit in the palm of your hand and solve certain problems septillions of times faster than this future dinosaur. To put that in perspective, a problem that takes Willow 5 minutes to compute would take El Capitan this many years. That's more years than the age of the universe. Even if you had an El Capitan supercomputer for every grain of sand on Earth, powered by a Dyson Sphere around the sun, it would still be many times slower than Willow. But the Americans aren't the only ones making breakthroughs in the quantum computing field. Just days ago, China introduced a record-breaking 504-qubit superconducting chip. After seeing this big number, people on the internet started freaking out because a computer this powerful could break unbreakable encryption algorithms and steal all those hock-to-a-tokens that you just yellowed your life savings into. But that's ignorant. By the end of this video, you'll understand how quantum computing actually works and why it still totally sucks as of today. It is December 10th, 2024, and we're watching The Code Report. On one hand, quantum computing could lead us to a utopia that unlocks fusion energy, artificial superintelligence, and immortality with nanorobot doctors. But it also poses an existential threat to humanity. Computing at this speed means that unbreakable encryption algorithms are suddenly rendered worthless. Any hacker with a quantum computer could instantly brute-force your RSA-encrypted communications on the internet. Or worse yet, crack your wallet's seed phrase and steal all your daddy tokens and NFTs. But despite Google and China making some big breakthroughs, this technology is still nowhere near being useful. To understand why, you first need to understand how quantum computing works. Classical computers, like the one you're using right now, rely on bits, binary 0s and 1s, to process information. Each bit represents a single state, like a light switch, and you can combine them together to do complex things, like stream this video around the world. Quantum computers also use 1s and 0s, but put them in something called a quantum bit, or qubit. The key difference, though, is that they can represent the quantum superposition of multiple 1s and 0s at the same time. Like, you can imagine a qubit as a box with a cat inside, and when you open it, there's a certain probability that the cat will be dead or alive, but you don't know until you open it. Mathematically, the equation looks like this, where a and b are probabilities, called amplitudes, that the qubit will collapse to a 0 or 1 when measured. And that means one qubit can represent a ton of information and compute in parallel. What's really weird, though, is that these qubits can become entangled, where the state of one qubit is directly related to the state of another, even though they're physically very far apart. And that weird behavior can be used to coordinate computations using quantum gates. They can do things like flip a qubit and create entanglement to provide a similar purpose to logic gates in classical computers, but do so by leveraging quantum mechanics magic. Sounds awesome, but the big problem is that qubits are extremely delicate, and they consistently produce errors, with some qubits having higher error rates than others. And that means they constantly need to deal with error correction to keep the thing stable. Not to mention, these chips need to be kept at temperatures near absolute zero to even work. So don't expect an Apple quantum phone anytime soon. Now, one thing that's special about Google's Willow chip is that it's able to find qubits with high error rates and reconfigure them on the fly, therefore reducing the overall error rate. And what's especially weird is that the error-corrected qubits get exponentially better as they get bigger, and Willow is the first chip to ever do that. Another big problem with qubits is that they need to be aroused into a state of superposition, and can only maintain that state for about 20 microseconds in the past. But on Willow, they've increased it by 5 times, up to 100 microseconds. That's still not a lot of time, but Google's next big milestone is to build a long-lived logical qubit. Willow only has 105 qubits, but can still perform certain calculations, like finding the prime factors in a large number, much faster than classical computers. Algorithms like Shor's have been around since the 90s to do this, but we've never had the quantum hardware to actually run them. In theory, when quantum computers get to around 2,000 qubits, they should be able to break widely-used encryption systems like RSA, using simple brute-force techniques that would take classical computers millions or billions of years currently. And what's scary is that the Chinese just released their own quantum chip with 504 qubits. That's impressive, but the real thing to watch here is the error rate. Once the error rate gets low enough, we'll hit that inflection point that allows quantum computers to scale up exponentially, and change the world in unimaginably good and bad ways. But this timeline has been all messed up since that fateful day on May 28th, 2016. And the best way to keep up as a developer is by hanging out with other developers on daily.dev, the sponsor of today's video. It's a completely free social platform that curates all the best developer content on the internet, and helps you connect with other like-minded people. Instead of scouring the bowels of Reddit for an update on your favorite JavaScript framework, you can rely on daily.dev to pull all the best content from over 1,000 top sources, to get all the best news, tutorials, and videos in one place. And when you install their highly-rated browser extension, staying up-to-date becomes an easy daily habit. Join over 1 million other developers on daily.dev by using my invite link on the screen. This has been The Code Report, thanks for watching, and I will see you in the next one.