It's also a place of immense scientific potential. A place where the most powerful tech company of the 20th century is engaged in a very expensive race with the most powerful tech company of the 21st century to make discoveries that could lead to nothing more or less than a wholesale reimagining of what computers are and what they can do.
And so, of course, it's also a place that startups and venture capitalists are watching closely. Very closely.
Google says it's achieved a long-sought-after breakthrough in the mysterious realm of quantum computing. IBM begs to differ. And two startups with dreams of taking on the pair of established powers have brought in more cash. Add it all up, and you get one of the 12 things you need to know from the past week:
1. Google's quantum leapLet's get the confusing part out of the way. What the heck is quantum computing?
The most basic building block of traditional computing is the bit, which is a portmanteau of "binary" and "digit." That makes sense, because bits can only have one of two values; these are usually represented as either a 0 or a 1. Bits make up a computer's memory. There are eight bits in a byte, so there are eight billion bits in a gigabyte. There are 256 billion bits in your 32-gigabyte smartphone. You get the idea.
A computer that size can store and process a lot of information. But it would take such a machine a very long time to accomplish certain tasks, like simulating complicated molecular functions, or cracking complex codes, or creating more accurate models of financial markets. If you're into that kind of thing.
Quantum computers, on the other hand, rely on the qubit, which is a portmanteau of "quantum" and "bit." A portmanteau within a portmanteau! Qubits are based upon a rather complicated concept called quantum superposition; for our purposes here, it's the principle that a quantum object can be in multiple states at the same time. A qubit doesn't always have to be either a 0 or a 1. Once it is measured, it will take one of those two forms. But until it is measured, it can exist in superposition as a range of probabilities between those two results.
It's the sort of stuff that can make your head spin. It's also an idea that's the basis for two of the most famous experiments in physics. One is the double-slit experiment, which has been around in some form for hundreds of years.
A simplified version: Electrons are fired one at a time at a wall. In between the electron gun and the wall is a screen with two parallel slits. If you fire the gun a few times and then look at the wall, you'll see a wave-like interference pattern. The electrons go through both slits, not one or the other. But if you set up a special camera and observe the electrons as they pass through the screen, you'll instead find two vertical lines on the wall. Now, the electrons act as discrete objects rather than probabilistic waves and only pass through one slit. The very process of observation changes how the particles act.
Despite the experimental evidence, it's an idea that some of the most famous scientists of the 20th century found ridiculous. That includes an Austrian physicist named Erwin Schrödinger. In a letter to Albert Einstein, Schrödinger explained the implications of quantum superposition. Say you put a cat in a box. Also within the box is an atom of radioactive material that has a 50% chance of decaying in a determined period of time, and a fatal dose of poison that's released if the atom decays. If you leave the box closed for that period of time, you can't know whether the atom has decayed. It exists in superposition; it both has decayed and has not. Thus, according to one interpretation, the cat too is both alive and not.
Schrödinger intended this thought experiment to show the impossibilities of such a theory. Many very smart people agree with him. But in the ensuing decades, many other serious physicists have concluded that it's actually an accurate conception of how the quantum world works.
That was a bit of a tangent. But the uncertainty that both examples describe is probably helpful in understanding quantum computing. Two bits together can be represented as either 00, 01, 10 or 11. Thanks to superposition, though, two qubits can store all four values simultaneously. Three qubits can hold eight different values. And so on, exponentially, ad infinitum. A relatively small number of qubits can have incredible computing power.
I've got one more mind-bender before we get into the actual news. Quantum computers also rely on the idea of quantum entanglement, another complicated concept that basically boils down to the fact that the fate of certain particles can become intertwined. If observers know the state of one entangled particle, they can instantly know the state of its entangled peers, no matter how far apart those particles may be. It's a potential loophole to the speed of light's function as a universal speed limit. Einstein poetically described it as "spooky action at a distance." If qubits are entangled, measuring one allows computer scientists to know the state of the others.
Phew. OK. Anyway: This week, Google published a paper claiming it's achieved a true quantum computing breakthrough: Using a machine made up of 53 functioning qubits, the company performed a calculation in three minutes and 20 seconds that it says would have taken the world's best supercomputer 10,000 years to complete. Thus, Google says it's achieved "quantum supremacy," which is actually a real scientific concept—building a quantum machine that can perform feats that are impossible for traditional computers in a realistic timeframe—no matter how much it sounds like something a Bond villain would be concerned with.
Over the past several years, IBM has been one of several companies and nations battling with Google to reach quantum supremacy first. Big Blue didn't take too kindly to Google's claims this week. In fact, it called BS. Two days before Google officially published its paper, IBM put out a blog post saying that a current machine could in fact complete the calculation in two-and-a-half days, not 10,000 years. Which would take a bit of the shine off Google's achievement.
Either way, though, the move could herald a new era in computing. In an interview with the MIT Technology Review, Google CEO Sundar Pichai compared it to the Wright brothers' first flight in Kitty Hawk. In itself, perhaps not the most impressive feat. But it's proof that the general idea works. Airplanes can fly. And super-powerful quantum computers can be built.
The implications for a quantum computing revolution could be immense. Almost any industry that relies on highly complex calculations and simulations could be totally transformed, ranging from finance to weather forecasting, from quantum physics to AI. Experts believe such machines could be used to better model human chemical and biological functions and create new medicines. They could make current encryption practices obsolete. And, as with any frontier technology, there are surely untold potential uses that nobody has dreamed up yet.
But the dreaming is already fully underway. A day before Google's big announcement, a quantum computing startup called IonQ raised a $55 million funding led by Samsung Catalyst and Mubadala, with several other corporate and traditional VCs also participating. IonQ is building its own quantum computers that it plans to make available to other companies via the cloud; earlier this year, it announced a major breakthrough of its own.
And on Monday, a German startup called HQS Quantum Simulations brought in €2.3 million (about $2.6 million) in seed financing to fund its intriguing aims. Essentially, HQS wants to use quantum computers to run simulations that can discover new materials and substances with commercial potential. In a press release revealing the funding, the company cited better batteries and more efficient solar cells as two potential applications.
It's the sort of moonshot stuff that VCs dream of. And it's only the beginning. As Google, IBM, Microsoft, the Chinese government and the other players in the great quantum computing race continue to make incremental progress, expect a flood of startups capitalizing on their technological breakthroughs to emerge.