Quantum technology is anticipated to address challenges that classical computing cannot. However, it is also seen as a threat to security, particularly to cryptographically encrypted data.
To clarify some of the concerns surrounding the technology, Peter Shor, Professor of Applied Mathematics at MIT and the author of the well-known Shor’s Algorithm, shared his insights during a fireside chat organised by DigiCert.
“People have come up with various intuitive explanations of how the quantum world works, which are called interpretations of quantum mechanics. They all give the same prediction for what will actually happen, but there are very different ways of thinking about it,” he said.
Quantum computing at a glance
As a background, Shor differentiated between a quantum computer and a classical computer.
“In a classical computer, you start in a specific state. Then there’s a rule that takes you to the next state of the computer, and then the next state, and so on. You just keep on going until your computation tells you to stop, and then, hopefully, you have the answer,” he said.
In contrast, quantum computation allows the computer to exist in a superposition of two different states, whereby the quantum computer operates as a quantum system.
“You can make these superpositions of states interfere with each other, and that way, you can do computations that are not possible on a classical computer,” Shor added.
To illustrate how quantum mechanics works, Shor described the ‘Many Worlds Interpretation,’ which suggests that whenever an action results in two outcomes, the world splits into two, creating all the many worlds.
“One way of thinking about this is that it’s not actually the world that splits in two, but you. You exist in a state entangled with the outcome of the experiment. When the experiment says one, the entangled state of you also says one. When the experiment says two, the entangled state of you says two. You are therefore in a superposition of two different states of your brain. This is really very counterintuitive and hard to think about, which is why a lot of people don’t like it,” he explained.
In contrast, the ‘Copenhagen Interpretation’ proposes that quantum mechanics consists of a sequence of unitary evolutions, where changes occur gradually. When a measurement is made, the world instantaneously shifts to reflect the result.
“This is really very difficult to understand intuitively, but it’s very useful in terms of thinking about quantum mechanics,” Shor said.
Looming threat
Published in 1994, Shor’s Algorithm enables the factorisation of large numbers almost instantaneously, solving equations much faster than classical computation. Fast-forward to 2024, and quantum computing is already being used to tackle complex enterprise challenges. However, the anticipated arrival of a cryptographically relevant quantum computer (CRQC) is raising serious concerns within the security community.
A recent study by Australian researchers warned that cryptographically encrypted data could soon be decrypted at scale by malicious actors if no immediate steps are taken to secure IT networks against “quantum hacking.” Similarly, Curtis Simpson, Chief Information Security Officer at Armis, has cautioned that the next generation of quantum computers will render traditional encryption algorithms obsolete, compromising sensitive data.
“In the wrong hands, quantum computers will be able to decrypt anything — from personal files and professional data to trade secrets and national security plans. The potential consequences are far more severe than those predicted for Y2K, and unlike Y2K, there is no specific deadline for when we must be ‘ready,’” Simpson said.
According to Shor, building a fully functional quantum computer remains a highly challenging task.
“In 1994, Jeff Kimball, a physicist, said that the state of the art in experiments allowed us to achieve only 1/10 of a quantum gate. To factor relevant numbers, you need to perform millions of quantum gates, and you also need thousands of qubits. Therefore, we needed new methods for building quantum computers. Right after I published my algorithm, Ignacio Cirac and Peter Zoller proposed quantum computers based on ion traps. This was a very clever idea, and it remains one of the primary architectures for quantum computers today,” he noted.
Shor also acknowledged the severe security implications if a functioning quantum computer were available today.
“Chances are that the first quantum computer will be very slow, very expensive, and capable of cracking only five or 10 RSA codes per day. It wouldn’t be used to crack the code you use to buy things online, because there are far more lucrative codes to crack. Even so, there are some very important secrets protected by RSA. I don’t know how much the NSA (National Security Agency) would pay to have those codes broken, but it wouldn’t be cheap,” he remarked.
Back to the drawing board
Asked about the possible timeline for cracking RSA, Shor said that it cannot be predicted at this time.
“It’s really going to depend on how many breakthroughs we get in the next few years,” he said.
On a brighter note, Shor sees quantum computing addressing abstract physics problems in the foreseeable future, as well as some industrial challenges.
“At the moment, we do not know how high-temperature superconductors work, and it’s possible that we could experiment with quantum computers and figure out how they work. If we do that, it’s also possible that we could develop better, high-temperature superconductors. A high-temperature superconductor that works at room temperature would have an amazing impact on everything,” Shor explained.
Quantum computing could also be applied to optimisation issues, he added. These include problems like supply chain logistics, financial modelling, and resource allocation — areas where quantum computing’s ability to analyse complex scenarios could significantly benefit enterprises.
“Most optimisation problems are completely irrelevant to practice, but there are some optimisation problems which companies would find very profitable if they could solve them. If quantum computers could solve any relevant optimisation problems, then that would be very useful,” he concluded.
