Networked quantum computers will solve unimaginably complex problems, says Gemma Church – but we need to secure the links first.
Quantum computers are coming, and that’s both an opportunity and a threat.
An opportunity, because quantum computers are vastly more powerful than conventional computers – the kind of machine that you’re probably using to read this article right now. That means quantum computing has far-reaching implications, allowing us to solve certain problems much faster than traditional computers – including some that can’t currently be solved at all.
In the right hands, quantum computers could unlock a range of breakthroughs in everything from medical science to climate change. To make the most of that potential, we will need not just to use quantum computers in isolation, but to network them – just as the usefulness of conventional computers really became apparent once they joined the internet.
“Before the internet, we thought desktop computers served no purpose outside of the world of research. Now look where we are,” says University of Innsbruck professor Tracy Northup, part of the pan-European Quantum Internet Alliance project. “The problems a network of quantum computers could resolve in the years ahead will be highly relevant for our daily lives and future progress.”
That’s also why quantum computing is a threat. One of the problems it could resolve is how to crack the encryption that protects our confidential information online – including everything from your email password to your credit card details. To hack an encryption key with a conventional computer is impossible in any practical sense. But in as little as 10 years’ time, hackers could use quantum computers to determine an encryption key in minutes.
The good news is that solutions exist to both effectively network quantum computers and to fend off malicious quantum hackers. Even better, it’s the same solution to both: an upcoming technology called quantum cryptography.
Quantum computers could unlock a range of breakthroughs in everything from medical science to climate change.
Every day billions of people transmit sensitive information on the world wide web. You might log into your email account, make an online purchase or trade private messages on a platform like WhatsApp. Every transmission carries a certain level of risk, but you carry on regardless, assured that your information is safe thanks to a host of tried and tested protective measures.
Digital encryption is one such measure: your data is encoded to protect it from prying eyes as it passes between two locations, using a digital key. If someone managed to grab the data in transit, they would need the key to unscramble it. Keys are notoriously difficult to hack because they are based on “one-way” mathematical functions. For example, if you multiply two very large prime numbers together, it’s practically impossible to find those prime numbers from their product using a traditional computer. That would require a vast amount of computational power and take an unfeasibly long amount of time to complete.
But tomorrow’s quantum computers will complete certain computational tasks many orders of magnitude faster than today’s computers, including such factorisation problems, because they operate in a fundamentally different way.
A traditional computer operates on bits of information: these bits are either a zero or a one, but never both. A quantum computer, on the other hand, uses quantum bits, or qubits. Qubits can represent a zero or a one but they also exhibit a property called superposition, which means a qubit can achieve a mixed state and exist as a zero and a one at the same time. Because of this superposition, a single qubit can represent two states at once. A second quantum property – entanglement – also means that multiple qubits can be linked together.
As a result, a quantum computer using qubits can represent many more states than a conventional computer using the same number of bits, allowing it to solve many problems far more quickly.
Let’s look at a simple example to demonstrate this point. If a regular computer has two bits, there are four possible states (00, 01, 10 or 11). Imagine these four states are four doors and behind one of those doors, there’s a prize. With a traditional computer, you have to sequentially open one door after another until you find the prize. With a quantum computer, you could open all the doors at once and claim your prize almost instantaneously.
Quantum computer using qubits can represent many more states than a conventional computer using the same number of bits, allowing it to solve many problems far more quickly.
As a result of superposition and entanglement, every qubit added to a quantum computer increases the number of states by a factor of two. So, two qubits can represent four states simultaneously, and ten qubits can represent 1,024 states, and so on.
Although forms of encryption that would offer some protection against such quantum attacks have been proposed, the odds of them being universally deployed are vanishingly slim. And any information previously encoded with a vulnerable form of encryption would remain vulnerable. So, a hacker could capture encrypted messages today and then use a quantum computer to decipher them when the technology has matured. That leaves the online information of billions of people open to attack.
This is where quantum cryptography is needed, in the form of a promising technology called quantum key distribution (QKD). QKD is a step-change compared to today’s encryption techniques because its security is not based on mathematical puzzles. Instead, QKD is based on the laws of physics: specifically, the rule that as soon as you try to observe a particle in the quantum world, its state changes, with its original state being lost.
QKD exploits this to make a message incomprehensible to anyone other than the intended recipient. Under QKD, the encryption key is generated as a series of encoded photons. If someone tries to read these encoded photons, their states will change, alerting users that their communication is no longer secret. That means it is in theory un-hackable – or at least, that any hacking attempt would be readily detectable. If no hack is detected, the security of the key is assured and encrypted communications can proceed without fear of eavesdroppers, allowing the transmission of highly sensitive data.
Over the last decade, the adoption of quantum cryptography has started to build, with numerous government and commercial institutions installing QKD systems for secure data transmission. But most of these installations only work over short, metropolitan distances, of the order of tens of kilometres. This is because the photons in QKD can only be sent so far before they scatter and lose their quantum state. This means data rates drop with increasing distance.
Thus, the information has to be decrypted and retransmitted periodically, which requires a high-security installation and some “trusted” intermediate nodes which act as relays between QKD links. At these nodes, keys are decrypted into their classical form and then re-encrypted in a fresh quantum state for the next section of their journey.
While the foundations of quantum cryptography make it theoretically un-hackable, vulnerabilities could be created as it moves out of the laboratory and into the real world. But the world’s quantum experts are investigating a range of options and remain confident that they can implement these techniques securely. For example, standardisation would introduce a robust certification methodology that bridges the gap between theoretical proofs and practical implementations with imperfect devices.
Professor Stephanie Wehner at QuTech, Delft University of Technology explains: “With the quantum internet, at least we are starting with something that is theoretically completely secure. In contrast, in classical computing, we started from a place that was not completely secure and also experienced further vulnerabilities in implementation.”
We’ve seen tremendous progress in QKD over the last few years. Increasingly advanced network demonstrations have been completed in the US by defence agency DARPA, in Europe under the SECOQC initiative, in Japan for the Tokyo QKD Network consortium, and by others. The joint China-Australia Quantum Experiments at Space Scale (QUESS) satellite mission also recently linked two laboratories 1,200 kilometres apart, demonstrating the possibility of intercontinental quantum transmissions. Last year, the UK and Singapore announced a £10 million collaboration to develop a new QKD test bed and push the transmission lengths of this technology further. Testing has also just begun of the world’s first commercial-grade quantum link between Suffolk and Cambridge in the UK.
The cost and practicality implications of quantum communications also need to be more fully understood, including the full life cycle support costs for commercial QKD systems. Depending on how quantum cryptography is implemented, it may require an entirely new global network or a major upgrade of existing networks. And, as with any new technology, hardware costs will need to come down substantially to realise widespread adoption.
The untapped potential of quantum cryptography means that, as this technology develops, it could have repercussions beyond the provision of robust security safeguards. Quantum communication could unlock a host of applications that simply are not possible using today’s networks, across a so-called “quantum internet”.
“Many years ago, I realised that QKD was not just a beautiful idea but something that can be implemented and used for good in the real world,” says Professor Nicolas Gisin, a co-founder of ID Quantique, which sells QKD and other quantum-secure systems. “There is a huge role for quantum cryptography to play going forward because it is the foundation on which every other potential quantum application is built.”
In truth, the full scope of these applications is still unknown. “As with any radically new technology, it is hard to predict all uses of the future quantum internet,” some of the world’s top researchers in this area wrote in a recent review article in Science. “However, several major applications have already been identified, including secure communication, clock synchronisation, extending the baseline of telescopes, secure identification, achieving efficient agreement on distributed data, exponential savings in communication, quantum sensor networks, as well as secure access to remote quantum computers in the cloud.”
So, what exactly is a quantum internet? Well, it is simply a computer network: in a similar vein to today’s conventional networks, computers are linked and pass information between one another. While it’s possible to connect quantum computers via conventional connections (and to conventional devices), their full power will only become apparent if there are true quantum connections between all the nodes on the network. So, with a quantum internet, quantum bits (qubits) are passed across the network.
This may sound like the stuff of science fiction, but work is being carried out to establish the world’s first quantum internet, which will connect four cities in the Netherlands by 2020.
Because qubits are creatures of the quantum realm, we can capitalise on entanglement: different qubits have correlated quantum states even if they’re separated by a great distance. (This is what Albert Einstein famously labelled “spooky action at a distance”.) And if the world’s quantum computers are entangled, using quantum connections, the resulting mass entanglement would create a global quantum computing cluster where information is effectively teleported between each node.
When we say ‘teleported’ in this context, we’re not talking about Star Trek-esque technology. Under quantum entanglement, if a pair of photons are entangled and one of the particles in the pair interacts with a third particle, the quantum information from this third particle is transmitted across to the other particle in the pair, effectively taking on its identity. It’s as if the third particle has travelled from one location to another. There is thus no real distinction between the different quantum computers in the cluster - just one enormously powerful quantum computer.
This may sound like the stuff of science fiction, but work is being carried out to establish the world’s first quantum internet, which will connect four cities in the Netherlands by 2020. This network will transmit qubits between any two network nodes, each of which consist of a few qubit processors. It’s hoped that the network will be available to external researchers to run test programs by 2021.
This is a significant development compared to the quantum links that are already up and running. Stephanie Wehner is working on the project: “Building a quantum internet is a long term and complicated project. Unlike the classical internet, where everything is well-defined, we are trying to do everything at the same time. We need to develop the hardware, the software and all the protocols and interfaces to make the quantum internet work.”
Quantum computing, quantum key distribution, and the quantum internet are all tricky to understand and even trickier to build. But the upshot is that if we can link up networks of quantum computers, then the world’s research community will have access to a monumental amount of computational power “in the cloud”, without needing to physically own a large and expensive quantum computer.
Let’s say you’re a scientist that wants to explore the properties of a new material or medicine. Instead of going into the lab and conducting an expensive, lengthy series of experiments, you can now log into a quantum computer and run a quick simulation to test your ideas.
You could also safely share high volumes of data across a quantum network. For example, sensitive medical images, records and genomic information could be securely transmitted between clinicians, opening the door to seamless and data-rich global medical research. Quantum cryptography will also allow scientists to test their ideas using vast networks of powerful quantum computers without giving up their valuable IP.
Quantum cryptography holds the keys to a future where quantum networks will protect our online interactions and expedite tremendous progress across every industry and research field.
A quantum internet could also improve the resolution of the world’s most powerful telescopes, allowing scientists to probe some of the most interesting phenomena in our cosmos, including supermassive black holes, to answer some of humanity’s most fundamental questions about our existence and the evolution of the universe.
From security to the stars, the true extent of the impact a quantum internet on society can’t be predicted, just as we didn’t understand the implications of the internet when it was first launched 30 years ago. But it is clear that quantum cryptography holds the keys to a future where quantum networks will protect our online interactions and expedite tremendous progress across every industry and research field – pushing the pace of human discovery further and faster than ever before.
Gemma Church is a freelance science and technology writer, with degrees in physics and astrophysics. She is a former software developer, and began her writing career 15 years ago with Bluesci, Cambridge University’s science and technology magazine.