Technology for dummies, chapter one: quantum computing
Understanding the potential of what might be the next major technological breakthrough requires looking into the mechanics behind it. Doing so with no scientific background at all requires a lot of curiosity.
Being an economics and business student, previously a politics one, currently interning as a VC analyst, you can imagine my face when my manager said: “So, in the next month, I would like you to prepare a presentation on quantum technology and the related investment opportunities”. Quantum what?
Getting deep into a technology based on physics — a class I stopped taking 6 years ago, and quantum physics above that, seemed like an intellectual challenge even harder than figuring out my accounting lessons.
“Don’t worry, you won’t have to understand quantum physics”. Wrong. I had to. Or to try, at least.
I have come to understand that part of a VC’s job is learning about a technology and being able to judge its potential quite fast. And in this process, I regularly find myself searching for simple explanations of complicated technologies on Internet.
Given this experience and my reflection upon the topic, I want to provide a simple explanation of what quantum computing is, which can help anyone understand what newspapers mean by ‘Google claims to have reached quantum supremacy’[1].
It is important to note that the most famous saying about quantum physics is “the less you seem to understand it, the closer you are to getting it”. My advice is: take a mental step back before reading the rest of this article.
First things first, the main idea is that quantum physics describe the behavior of fundamental particles, how they interact and what properties are linked to their behavior. Two properties of quantum physics are central to quantum technologies: entanglement and superposition. These properties and the fact that we can (potentially) control them are what define quantum technologies’ capacities. For this article, I will focus on superposition, which is the base of quantum computing capacities.
Sit tight, this is where it gets a little twisted.
Superposition
So, a regular particle has two possible states. Traditional computers’ particles, the bits, are either in a state called 1 or another called 0. These particles, the bits, are binary: they are either 1 or 0. But quantum particles, or qubits in the case of computers, have more ‘fluid’ identities. Indeed, they can be either 1, or 0, or any combination of 1 and 0. This means that they could be 30% 1 and 70% 0, or any other combination.
That is what we call superposition: a superposition of both states, expressed as a percentage. Superposition means that a quantum particle can be in any of these states at any time.
However, when measured, a quantum particle will always be 1 or 0. But before measurement, it could have been at any state between 0 and 1. Personally, I imagine a little red and blue ball constantly changing color between red, purple and blue, which stops changing color once you pick it up and becomes either blue or red.
But scientists, much smarter than me, have depicted the concept in other graphical ways, and the most common is Schrodinger’s illustration with a cat. Schrodinger imagined a cat, trapped in a closed box, where there is a mechanism likely to kill the cat. The probability that the mechanism kills the cat is one out of two, or 50%.
The idea is that as long as the box is closed, you do not know whether the cat is dead or alive. Same as before measuring a quantum particle, you do not know whether it is 0 or 1, red or blue. As long as you have not opened the box, the cat is both dead and alive: it is in a state of superposition.
Still following? Good.
Now that we have established that a quantum particle can be in two states at the same time, we can say that a qubit is thus both 1 and 0. And this is what gives a quantum computer its power.
A regular bit is either 1 or 0, and thus N classical bits contain an amount of information equal to N. A qubit, on the contrary, being both 1 and 0, N qubits contain an amount of information equal to 2^N. Hence, the information contained in a quantum computer is exponentially bigger than in a traditional computer, providing it with an exponentially more powerful computational capacity.
The technical part stops here. The main idea is that quantum computers, thanks to quantum particles, possess a computing power that is much (much) bigger than that of a classical computer. It can solve a much (much) higher number of problems.
Now, it does not mean that quantum computers will replace traditional ones. The computing and calculating power of quantum computers is much bigger, but it does not mean that it will be capable to do everything much faster. On the contrary, it might even take a little more time to solve problems. It is not a matter of time, but of capacity, of scale.
Do you begin to imagine the applications of a computer that can solve a lot more problems at the same time than a traditional one?
There are multiple sectors of application in the foreseeable future: any sector with an organizational aspect to it or any that requires great calculation power.
What could a quantum computer do?
I use the conditional here because quantum technologies are still very early in their developments. Google has, indeed, managed to complete calculations with its quantum computer that could not have been completed by a traditional computer — quantum supremacy. Startups in the sector have raised tens of millions of dollars of VC funding. But most technological advances are still limited to the embryo stage and do not have direct applications today.
IBM [2]distinguishes three main capacities for quantum computing: machine learning, simulation and optimization.
Machine Learning
Quantum technology and AI could actually help each other: training a Machine Learning algorithm on a quantum computer could make it much more powerful and accurate, while basing quantum calculation on a self-learning and constantly improving AI model could empower quantum computers’ calculation mechanisms.
A virtuous circle.
Optimization
Optimization is the idea that any industry that relies on complex organization or calculation could be enhanced by quantum computing, which can solve numerous problems at a time.
Quantum optimization would allow avoiding the shortcuts used by traditional computers for problem solving, minimizing the risk of errors while enhancing the process. In application, this would improve any infrastructure, network or system based on organization — think class schedules, air traffic, shipments, telecommunications networks…
In finance, this could imply improvements in trading strategies, portfolio optimization, asset pricing and risk analysis. J.P. Morgan Chase is currently testing such hypotheses on IBM’s quantum computer.
Simulation
The final one, simulation, is my favorite. Quantum algorithms are expected to have the capacity to compute accurate simulations of molecules that are impossible to model precisely with our current computers. IBM scientists have been testing the simulation of large molecules on a quantum computer, showing promising results.
To find the medication for a disease, you need to compute the molecule that will cure the disease. However, today, our computers do not have sufficient calculation power to do so. But, a quantum processor could do it. An example given by IBM is the following: “a future quantum processor could simulate a caffeine molecule — this would require a gargantuan conventional computer larger than 10 percent of the size of the earth” [3].
Basically, a quantum computer could potentially simulate the right molecule to cure cancer, or Alzheimer’s.
This application is also the one predicted to become the one driving quantum’s market growth in the following years.
Startups researching such solutions have attracted the majority of VC funding in quantum, such as ProteinQure, which develops a computational protein drug design platform, and raised $4M in July with Global Founders Capital, Inovia Capital, Golden Ventures and Felicis Ventures.
A final application is security, and particularly the encryption of data and communications.
Security
The encryption of data and communications is one of cybersecurity’s central focuses today. And the encryption of data through quantum technologies has been identified by CB Insights as a trend that is still in the early-adoption phase and has a large addressable market and the potential to revolutionize it.
Encryption through quantum is based on quantum cryptography. In cryptography, both sender and recipient share a key to decode the sent message. In a quantum world, this key is based on a concept called Quantum Key Distribution (QKD). QKD is the fact of encoding information in qubits, using their quantum properties, such as superposition.
The idea is that the encryption, the code, is based on the fact that measuring a quantum particle changes its state: it is no longer in superposition but becomes 1 or 0 (remember the red and blue ball). If a communication encryption is based on quantum particles, it is based on particles of unknown states, which change at any attempt to measure them. And because an attempt to decrypt the message, to eavesdrop on the communication, would require measuring the qubit, any interference would alter the key and alarm both senders and receivers.
This means that communications would be ‘un-hackable’ and most importantly it would allow measuring the amount of information that has been intercepted.
Large telecommunications group have started looking into quantum properties for communication, such as SK Telecom, which is applying quantum cryptography to the deployment of 5G, or Huawei, which is experimenting applying quantum cryptography on commercial optical networks for secure communications with Telefonica.
Startups like Isara or Post Quantum, which offer post-quantum encryption-based data and communication security solution, and have raised a combined total of $33M in the past three years.
The field of possibilities for quantum technology applications seems vast and very promising. A world with cures for cancer and completely optimized financial investments, with no data breaches and automatized everything would surely be an improved version of our world. But the technology remains still very young, and its potential applications based on ‘could do’.
Now that you understand — I hope — a little more what quantum computing and technologies are all about, I let you decide for yourself whether these ‘could do’ are a matter of months, years or decades.
To hear an inspired and inspiring quantum physicist explain the principles I depicted above, I encourage you to watch this TED Talk by Shohini Ghose: https://www.ted.com/talks/shohini_ghose_quantum_computing_explained_in_10_minutes?language=en
To further document yourself on the topic, and its possible applications:
· IBM (2018) “Coming soon to your business — Quantum computing”
· BCG (2018) “The Next Decade in Quantum Computing — and how to play”
· New Scientist “Google has reached quantum supremacy — here’s what it should do next” https://www.newscientist.com/article/2217835-google-has-reached-quantum-supremacy-heres-what-it-should-do-next/
[1] Financial Times (2019) — https://www.ft.com/content/b9bb4e54-dbc1-11e9-8f9b-77216ebe1f17
[2] [3] IBM (2018) “Coming soon to your business — Quantum computing”
