You’ve probably heard of Moore’s Law, but in case you haven’t: Moore’s law (created by Intel founder Gordon Moore in 1965) states that the number of transistors per square inch on a computer chip is likely to double every two years, while the costs of developing these chips are halved.
So, what’s a transistor and why does this matter? In the most basic sense: a computer chip has some elementary modules (or functions), which run logic gates (and, or, not, etc.), which are powered by transistors. This sounds complicated, but we can use a straightforward example to clarify things. A simple calculator has a computer chip which has functions (or modules) like add, subtract, multiply, etc., and these modules are calculated using logic gates. A transistor is what translates the number 3 into binary, or a combination of 1’s and 0’s (known as bits) which is the most rudimentary computational language used by CPUs. So why do we use binary? Because electronic computers use electricity to represent real information (letters, numbers, images, etc.) and electricity can either be sent or not, represented by a 1 or a 0. Now that we’re all professionals on how computers work, back to Moore’s Law:
The exponential growth described in Moore’s Law was originally applied to computer chips, but similar trends exist in applications such as computer memory, digital camera pixels, and the resolutions of displays and streaming services. However, exponential growth is typically unsustainable, and many people think that we’re reaching the end of the line when it comes to computational capacity. In 2015, the former CEO of Intel stated that Intel’s “cadence was closer to two and a half years than two”, indicating a slowdown of Moore’s Law. And this makes perfect sense. After all, even if we are able to make transistors that are the size of individual atoms, we will still reach a point where we can’t cram any more computational capacity into a computer chip. Today’s transistors are around 14 nanometers big, which is 500 times smaller than a red blood cell. However, even at 14 nanometers, the transistor is still big enough to prevent electrons from passing through. But at a certain point, the transistor becomes small enough that the electrons can pass through it using a process called quantum tunneling.
Here’s where quantum computers come in. Quantum computers take advantage of this process to expand on the capabilities of bits. Instead of simply having an electron with a value of 1 or 0 (sent or not), quantum computers use photon-based qubits that can exist as a 0, 1, or any proportion of the two (for example, 50% – 1 and 50% – 0), depending on their polarization. To picture this, imagine you’re wearing polarized sunglasses and looking at your phone screen. As you turn your head, the image darkens, since the lenses are polarized to an angle perpendicular to the light waves from the screen. Keep turning your head a full 90-degrees and the image becomes visible as the lenses are polarized to an angle parallel to the light waves.
Now here’s where things get a little confusing (yes, NOW they get confusing): when a qubit is observed (i.e. passes through a filter) it has to ‘decide’ if it is a 0 or 1 (horizontally or vertically polarized). But until that point, a qubit exists as both a 0 and 1. You heard me right, it’s both – and this is called superposition. If we take 4 regular bits and they each have a value of 0 or 1, they can be in one of 24, or 16, configurations (0000, 0001, 0010, 0011, …, 1111) but only one can be used at a time. 4 qubits, however, can be in each of these 16 configurations simultaneously. At this rate, 20 qubits can store over a million configurations simultaneously. At this point, I’m sure you’re wondering “So what? Why is this important? Why should I care about some new type of computer that I’ll never get to use?”
The easiest answer is that quantum computing has the potential to completely change the way we think of cryptography. Everything you do online today, from checking your emails to sending bank account information, is protected through encryption. Encryption is a way of obscuring data so that, even if intercepted, it is impossible for someone to read the data without your private key. However, quantum computers should be able to solve encryption algorithms significantly faster than today’s computers (which take years). Even things that we consider to be un-hackable (e.g. Blockchain) will be susceptible to quantum computing.
However, this is just the easiest answer. Quantum computers can also be used to create models of quantum physics, helping us understand the fundamental structure of the universe. It can help us advance medicine at speeds we could never envision before by creating models of proteins that are too computationally heavy for today’s computers. They have the potential to enable AI algorithms to finally get us to the point that we can create general artificial intelligences. Quantum computers are also excellent for conducting searches of extremely large data sets. The classic example is a phone book with 100 million names, for which it would take a quantum computer 10,000 operations to find your name. Whereas with traditional computers, this would require an average of 50 million operations to accomplish the same task. The bigger the phone book, the bigger the gap between quantum and traditional computers. And in today’s era of big data and data centers, this type of efficiency has limitless applications.
But the technology is in such an infant state that I believe we won’t even begin to realize its true potential for many years. In addition to the barrier of qubits’ physical limitations (they require carefully shielding and temperatures near absolute zero to function), I think that many people won’t realize their value until they see them produce results. This is why so many companies (IBM, Google, Microsoft, JP Morgan, and more) and the US government (to the tune of $1.3 billion) are investing so heavily into quantum computing. They’re all running a race where the exact prize may be uncertain but undoubtedly comes in an enormous box.