On October 23nd 2019 Google announced in an article in the magazine Nature that its Californian Sycamore computer had managed ‘quantum supremacy’, using its 53-qubit processors for a calculation in 3 minutes 20 seconds which would have taken 10,000 years to do on the fastest classical supercomputer around today. In this case ‘supremacy’ means proving itself better than today’s classical computers. IBM has already disputed the claim, estimating that the same calculation could theoretically be solved in just two and a half days on its own Summit supercomputer in the Oak Ridge Laboratory in Tennessee.
Quantum computing is a hard subject to master, even for the scientists developing it; let alone for any if the rest of us to understand. There are many challenges to get the prototype machines to work and to prove that they’ve succeeded in their calculations. However it’s a subject we can’t ignore, because if and when they can be put to practical work they will enable us to solve problems unachievable with our current classical digitally-based ones.
The types of problems quantum computing – when working – will solve include modelling chemical and predicting reactions, as well as optimizing transportation and logistics systems. It will create a major step forward in the creation of true artificial super intelligence, although we haven’t really begun yet to think of the majority of the applications it can be put to. On the downside it will also be used to crack the encryption codes used to protect data starting with the simplest, which has stimulated much research into new ‘quantum-safe’ techniques.
All current computers are digital machines, converting all data into bits, which have only 2 binary states – 1 or 0. Quantum computers are based on qubits, which have many more (see the Bloch sphere in the Figure).
Quantum computing is currently being developed by computer companies IBM, Microsoft, Intel and Google; processor developers such as QuTech and IonQ and government institutions across the globe including the USA, China, the UK, Australia and the Netherlands. In fact Wikipedia lists over 70.
There are still many years to go before quantum computers can be used for practical computing problems. The processors have to be cooled to near absolute zero (why pictures of them show processors encased in glass) to overcome the errors of qubits. Error correction is one of the biggest challenges to surmount; perhaps Google’s greatest achievement was to overcome the failure of one of its processors – why the Sycamore calculation ran on 53 processors rather than the 54 it was built with. Quantum computers will not replace classical designs any time soon; not only are they useless at mathematical equations, they also require a mass of classical computing to input, correct errors and interpret results. They will, however, extend they types of problem that can be solved – especially those that are theoretically possible, but practically unachievable, today.
It is impossible to predict when future quantum computing breakthroughs will take place, although it is heartening to see the interaction between developers from different organizations – less secretive than the development of atomic science and the A bomb in World War II. Nevertheless the approaching arrival of practical quantum computing increases the need to address the ethics of algorithms to protect society from the potential catastrophe of super intelligence. The real ‘supremacy’ isn’t over classical computer architectures, but over the human brain!
©ITCandor Limited – unauthorised copying of this content is illegal and will be rigorously defended by us through court action