skip to navigation skip to content

Brain food


Quantum computers will change our lives – but how quickly?

In the first of a three-part series, we ask how quickly will quantum computers change our lives?

Hipster businessman with virtual screenQuantum physics? Check. Quantum mechanics? Check. Quantum computers? Erm, computers? That’s right – quantum computers. Millions of times more powerful than conventional computers, quantum computing could lead to huge improvements in machine learning, artificial intelligence, computer simulations and cryptography. All of which could fundamentally alter the way our society operates.

The essence of quantum computers’ enormous potential lies in the ability to perform a huge number of simultaneous calculations – far more than a conventional computer. While a normal computer is based on bits of information that exist in one of two states (0 or 1), in a quantum computer information is encoded as quantum bits (qubits), based on subatomic particles, and so capable of being in “superposition”, or multiple states of 0 and 1 simultaneously.

Ilyas Khan

Ilyas Khan

Sounds great, but how quickly will quantum computers actually get out of the laboratory and start to have a noticeable impact on society? Ilyas Khan, Leader in Residence and Fellow in Management Practice at CJBS, is certain we will soon see the numbers of quantum computers in use by governments and some private organisations rapidly increase. He believes a tipping point has already been passed in terms of government and private enterprise investment. “In the last three or four years,” he points out, “hundreds of millions of dollars have been spent [on quantum computing research] by people who’ve already made a difference to our lives: Google, Microsoft, IBM, Intel.” It’s a change, he says, that could lead to technological, scientific or medical breakthroughs that really could affect all of our lives.

Aram Harrow, Assistant Professor of Physics at the Massachusetts Institute of Technology (MIT), agrees. He thinks that quantum computers will be able to outperform conventional computers in fields such as machine learning (training computers to use data to, effectively, make decisions without additional human input, to run search engines, spam email filters, voice or facial recognition technologies, or self-driving cars, for example) and simulation technologies.

Khan and Harrow recently joined forces at a Town & Gown debate hosted by CJBS to argue that quantum computers are soon going to be affecting our everyday lives. Arguing the other side, in front of an audience of 300 and including guest of honour Stephen Hawking, was Dr Hermann Hauser, serial entrepreneur and co-founder of Amadeus Capital Partners. Speaking in favour of the debate’s proposition, that “quantum computing, while scientifically interesting, will not make a practical difference to society in our lifetime“, he pointed out that many obstacles continue to delay the development of commercially viable quantum computing.

Dr Hermann Hauser

Dr Hermann Hauser

Dr Hauser argues that existing conventional computing technology is mature and still developing rapidly, with hundreds of billions of dollars invested in these technologies every year; and that until quantum computing starts to attract a similar level of investment it stands no chance of replacing it as our main computing platform. “Instead,” he argues, “quantum computing will remain a specialised pursuit, possibly used only for the sorts of purposes for which conventional supercomputers are used today” – such as climate research or molecular modelling of chemical compounds.

He describes the problems still facing anyone attempting to create hardware or software for quantum computing. He points out that the current world record for a controlled entanglement of qubits – the mechanism that enables quantum computation – is 14 qubits. “For a quantum computer to challenge classical computing you need at least 50 qubits.”

In theory, says Dr Hauser, quantum computing could use Shor’s algorithm for integer factorisation (an algorithm that finds the prime numbers by which any given number is divisible), and so could crack the cryptographic codes used in most digital data encryption, which are based on the factors of very large numbers. That would make it much harder for governments, business or anyone else to protect electronic data. But in practice, he points out, it has not yet been proven that software for a quantum computer would be able to use Shor’s algorithm any more efficiently than a powerful classical computer.

Above all, he says, the expense and difficulty of developing quantum technology means that at present it is a long way behind classical computing in terms of commercial viability: “It is very difficult to produce a competitive product for, say, $1,000 that can beat existing technology.” He concedes that it will become easier and less expensive to develop quantum technology over time, but suggests it will still be a long time before these technologies can be provided as cost effectively as can conventional computing.

Also at the debate was Professor Jeremy O’Brien, Director of the Centre for Quantum Photonics at the University of Bristol, who points out that one reason quantum computing might struggle to impact everyday life is that it may be suppressed by those opposed to changes it might bring. That might include governments or businesses that don’t want to see encrypted data cracked, but also some white collar workers: “White collar jobs will disappear. That’s starting to happen already, with conventional artificial intelligence and machine learning. People working in the financial industry [could] be serious opponents to change.”

Although he lost the vote on the night of the debate, Dr Hauser agrees that quantum computers do have the potential to affect all of our lives at some point in the future, but remains convinced that the practical obstacles to their development will continue to restrict their widespread use for years to come.

But while Aram Harrow concedes that they might at first be used to run complex simulations such as those commonly run by conventional supercomputers, he points out that this may still have major consequences: “If there was only one useful new drug, or we only figured out … [how to] make better NMR and better MRI machines…. If we only made one new composite material that we could build cars out of – that would make an enormous practical difference.”