Quantum Computing: Hasten Slowly.

Every now and then, some idea or potential technology goes through a fair bit of hype. Like you, I have seen my fair share of hyped-up technologies. The “inflated expectations” as Gartner calls them often comes crashing down.? Sometimes things do take off, but the “slope of enlightenment” is a slow and gradual one.?

This is a problem, because those that fall victim to the hype invest in ideas, companies and projects that will more than likely go nowhere.?Intelligent people experience fear of missing out (FOMO) or fear, uncertainty and doubt (FUD). Few people call out the hype, as everyone likes a good idea, and who knows? Anything is possible. History proves many doubters wrong, if you wait long enough.?

I have seen and been a victim of many hype cycles, so thought I would call this one out early.

What is quantum computing?

Quantum computing is based on one of the most bizarre realities in the universe – the mysterious behaviour of quantum particles.? They can be in an indeterminate state, and act one way (statistically) when their state is unknown, and another way once measured.? Without solving the mystery, someone came up with the idea that it was possible to use this principle, and another associated behaviour, quantum entanglement, to create a computer.? Traditional computers even today are built on binary states.? A “bit” is either 1 or 0.? In quantum computing a bit, or “qubit” could be in a mixed state, both 1 and 0 with probability of either, until measured. This is actually quite amazing and perplexed Einstein, and other scientists, like Schr?dinger who illustrated the bizarreness in his Schrodinger’s cat thought experiment .

Intuitively you may think that a computer that didn't know what state it is in, is actually a worse computer, and you would be correct: for it turns out for most “classical” computing problems, a quantum computer will probably perform less efficiently than a normal computer.

Encryption

The story of the development of complex computing, centres ironically not on the need to share data in the information age, but rather to conceal information.? The first computers, the Polish/British Bombe of WWII was designed to break the German encryption machine, Enigma.? My friend who imported the first personal computers in Australia got into trouble with the US authorities because those PCs (like all PCs after it) had encryption capabilities above a US export limit.? In 1977, RSA (a company formed by 3 cryptographers) came up with perhaps the most revolutionary encryption ever: it allowed you to transmit and decipher secrets without sharing secret keys. This encryption system, and others liked it fuel the exchange of information on the Internet today, allow us to securely transact, and maintain our privacy.

RSA was based on the idea that it is difficult to factorise a large number that looks like a prime (a pseudoprime).? Factorising 21 is easy, because we have memorised multiplication tables. Very quickly it gets quite hard, take 332393 for example. For a computer, typical algorithms take time to work this out.? It will try and divide by 2, then 3, etc.? Modern encryption is 2048 bits, say 617 digits.? It would take longer than the age of the universe for all the computers on earth to factorise this, using conventional algorithms.? Intuitively, you may say that maybe it is because no one has come up with a good trick or algorithm to factorise primes?? You would be correct. In 1994, Peter Shor came up with an algorithm to factorise primes in less steps than any algorithm before it. The catch was it needed a theorised quantum computer.

So encryption using RSA worked on the idea that you could give the pseudoprime to everyone (the public key), but only the originator knew the factors (the private key).? A year after Shor’s algorithm, Netscape brought out SSL that used RSA (and later similar encryption), which started the HTTPS revolution that secured the Internet (the most common protocol used today to access the Internet). However, the weakness was already exposed before its birth. We were desperate for a solution, and this was the best thing around.

Development of the Quantum Computer

George Boole developed boolean algebra in 1847. It took over a hundred years to develop a working computer using these algorithms.? Similarly quantum computers were theorised in the 1980s … the journey to a real working and practical quantum computer has been a slow one.

Today, there are headlines of technology companies developing quantum computers with a large number of qubits.? IBM recently claimed Condor with 1,121 qubits. 2,000 theoretical qubits can do some damage, so a casual observer may think we are close.

Herin lies the gap. ?A quantum computer needs to maintain qubits in an entangled state. Larger and larger numbers of qubits get exponentially harder to maintain.? There is another problem: decoherence.? The act of measuring a qubit collapses its state, so it is no longer a qubit, but a classical bit. It is hard to know if the qubit has been measured, as interaction with the environment in the most innocuous way can cause decoherence.

This leads to a high error rate.? Traditional computers have errors in them. These are quite rare, and can be resolved by adding a redundancy, or check bits etc, so a user will generally never notice them. Quantum computers on the other hand have a 1% error rate today. This may mean that for practical purposes, one needs a lot more qubits to get the same accuracy. In practice, the largest factorisation using quantum computers has been relatively insignificant.

Also developing the circuitry, gates and logic elements of the quantum computer is another exercise.

?It will take quite some time before practical quantum computers become a reality.

Usefulness of a quantum computer

Besides Shor’s algorithm, and others in breaking encryption, there are few applications of quantum computing that have been proposed. Numerous hackathons and challenges have proved this. Complex optimisations may be one of them, better AI perhaps, but the examples are few, and far between. Build it and they will come, perhaps.

Conventional computers will continue to maintain their supremacy for typical applications.

Secret quantum computers and conspiracy theories

Some may suggest the US, Russia and/or China may be on the verge of secretly developing a sophisticated quantum computer, if they have not already.? Rarely has it been the case that major innovations in science have been kept secret for long. It took four years for the USSR to test their nuclear bomb after the US. If there was a power that was eavesdropping, they would need to be very careful on acting on that knowledge, the moment the cat was out of the bag, the advantage would be lost. So, the last thing they would do is steal our credit card numbers.

So what can we do?

First – treat any investment opportunity in quantum computing with a grain of salt. It is still in the research phase, and unlikely to generate a profitable outcome.? Universities and big tech corporations are investing in it to seed long term benefits, a public service, ?secure some patents, there may even be a Nobel prize or two as we solve some tricky problems in the coming decades.

The biggest commercially profitable pursuit right now is not in the development of the quantum computer, but protection against it. Uplifting encryption is especially important in my own industry, payments, where keeping money safe is essential. Increasing the number of bits of encryption, and better algorithms here is key.

For a user of encryption there is no need to rush, resist the fear, uncertainty and doubt.? Our popular encryption protocols such as TLS were designed to allow the upgrade of the encryption mechanism, and our providers are doing that just now.? Just keep upgrades set to “on”!

Consider also that one day, in the future, all our data will be exposed.? Blockchain, for example is an open ledger, it is just hard to read it without a key. All your transfers and data could become exposed in the next decades. One day we will know who stole which Bitcoin, and even legitimate transactions will be in the open. So consider storing sensitive information carefully, or not storing it at all. Tread carefully with Web3.0.

Is there a threat of disruption to your business if your competitor gets a quantum computer to solve specific business problems?? Currently we don’t think so, though certain complex optimisation challenges, like “the travelling salesman ” may be solvable with quantum computers.? Like AI it will take years to learn how to solve our problems with the new tech, so there is no need to panic.

Just as bacteria can evolve to resist antibiotics, newer technology can break old encryption, so it is good to think about the risks, and improve it as much as possible, before the theory becomes a reality. However, as an investor or user of quantum computing, don’t expect a practical model soon: the best advice is “hasten slowly”.

?