Common misconceptions and myths about quantum computing, hailed by some as the next great disruptive technology, emerged during an expert discussion that I introduced in the Science Museum at last week’s Lates. This event with UCL’s CompBioMed consortium was held to mark our latest blockbuster exhibition, Top Secret: From Ciphers to Cyber Security.
These next-generation computers rest on the principles of quantum mechanics, which was developed at the start of the 20th century to describe nature at the smallest scales of energy and matter.
Ordinary or so-called ‘classical’ computers were first put on a mathematical basis in the 1930s when English mathematician Alan Turing came up with a mathematical portrait of a universal computer.
Half a century later, it was pointed out that Turing was working with classical physics and that a universal computer based on quantum physics would have greater powers. This would mean that calculations that would take a classical computer longer than the history of the universe could be done in a reasonable amount of time.
The reason it would take so long is that classical computers encode information in the form of bits, which can take the value of 1 or 0 (think of them as the currency of on/off switches that ultimately control how a computer works).
However, quantum computers are based on ‘qubits’, which can represent both a 1 and a 0 at the same time (quantum mechanics refer to ‘superposition,’ where they are in states that are neither zero or one).
What is remarkable is that while a quantum computer with two qubits could run four calculations at the same time, a 1,000-qubit device could process more simultaneous calculations than there are particles in the known universe.
The rapt audience was told by its MC – theoretical physicist, author and broadcaster Prof Jim Al-Khalili – that, despite the strangeness of the subatomic world, quantum computing is increasingly making the headlines: ‘Are we approaching the time it is going to become a reality?’
During the event, various myths and misconceptions emerged:
Quantum Supremacy spells the end of normal computing.
News recently leaked that Google scientists had achieved “quantum supremacy”, where a quantum computer outperformed a regular computer on an especially hard task.
But the set task tends to be ‘extremely artificial,’ said Peter Love of Tufts University, who used to work with D-Wave-Systems, a Canadian quantum computing company. In the case of the Google work they tackled the ‘random circuit sampling problem.’
Although a quantum computer has won a race with a classical supercomputer, the practical implications are minimal, and it remains possible that classical algorithms can sometimes catch up.
Love added that when it comes to the much-disliked phrase ‘quantum supremacy’, ‘that’s a term that pre-dates Donald Trump.’
Quantum Computers can outperform normal computers.
Yes, for certain tasks, but no for other things at which they are ‘utterly useless,’ said Ian Levy, Technical Director, National Cyber Security Centre.
They will be useful to tackle hard problems like optimisation, such as how to make processes run more efficiently, whether finding the shortest routes for a delivery run, or providing search results from the internet, said Viv Kendon of Durham University, who works in the Joint Quantum Centre of Durham and the University of Newcastle. ‘They have several tricks that can bypass what classical computers can do.’
However, both are digital computers and Peter Coveney Director of Computational Science at UCL, and a chair at the University of Amsterdam pointed out in a paper published the day before that the numbers they rely on are a poor representation of reality.
So neither classical or quantum computers can completely reliably reproduce the behaviour of ‘chaotic systems’, which are commonplace.
Quantum Computers will be commonplace in 20 years.
Pundits have been saying that for the past twenty years, remarked Ian Levy.
Quantum Computers will replace normal computers.
Viv Kendon told the packed IMAX audience that computers are becoming more diverse and interconnected and the future lies in hybrid machines since classical computers remain very powerful: ‘We are not expecting quantum computers to sweep away classical computers.’
Quantum Computers will wreck internet security.
Security depends on hard problems, such as factorisation, finding the two prime numbers that must be multiplied together to produce a third. Using a classical computer to break a 2048-bit key would, based on his assumptions, take roughly the entire UK electricity generating capacity 150,000 years, said Levy.
However, quantum computers with around 100,000 qubits (we can only manage around 50 at present) could do this job in seconds. Ian Levy said that this is ‘kind of cool but really scary’ but added that mathematics already provides new ways of doing cryptography that cannot be cracked so easily by a quantum machine.
Quantum computers can run the same programs that we can use on classical computers.
They must use different kinds of algorithms and, to develop them, computer scientists need more quantum computing hardware ‘to try ideas out’, said Viv Kendon.
Quantum Communication will be important.
Quantum theory, which is deeply counter-intuitive, states that particles can be in two places at once, another example of superposition. Moreover, two particles can be related, or ‘entangled,’ so that their properties are linked, regardless of their relative distance, which some have used for secure communications since they reveal if someone is tampering. But these only provide links between given points, which is highly restrictive, said Levy.
Quantum computers will be better at modelling the real world.
Yes, when it comes to quantum systems, for instance when it comes to designing materials bottom-up, from scratch, a long-held dream described by Peter Love: ‘We really believe quantum computers can solve that problem.’
However, he added that although classical computing has failed to crack this problem for the past 60 years, scientists lack the proof that they will never be solved this way.
They will be used to help create a ‘digital doppelganger’ of patients, a so-called virtual human, to test treatments and drugs, said Prof Coveney. However, these simulations of the body will have to work at a range of scales, from the level of atoms (amenable to quantum computing) to the level of organs, where conventional computers can be used.
Quantum machines could become conscious.
Prof Al-Khalili warned the audience before opening the discussion that anyone who suggests that quantum computers, which are somewhat mysterious, can somehow solve the consciousness, itself a notoriously hard-to-define concept, ‘will be taken out and shot.’
Join more discussions like this at our monthly Lates event.