Powerful Quantum Computers | Today’s Data Security Threats

This mattered because factoring large prime numbers formed the basis of the difficult mathematics underpinning public key encryption (PKE) upon which web HTTPS, email, and cryptocurrencies have come to depend. A quantum computer would break this in seconds.

No such computer existed at the time but research into building one had already begun. It was clear that quantum computers based around quantum qubits (the quantum equivalent of a bit) would one day exist. Today, several companies have built working quantum computers although they remain experimental devices limited to a few dozen qubits, too modest to threaten PKE. One limitation of quantum computers is that the same physics that make them mathematically powerful also make it incredibly difficult to extract the answers they provide.

But with 1,000-qubit quantum computers promised for the near future, it’s clear this won’t always be true. Every year, incremental advances happen, edging closer to what some have called the ‘quantum apocalypse’.

This sounds a bit hyped so is this just another Y2K scare story? And do quantum computers have any implications for the average UK company today?

Harvest now, decrypt later

One worry is that quantum computers might undermine PKE security suddenly and unexpectedly. It’s most likely this would be done by a single private company or a nation state, but the effect would be the same – what had been secure one day would be open to doubt the next. This could be hugely disruptive. Because nobody knows when this will happen, this creates uncertainty. It might be five years away, or 10, or 15.

But there’s a second and even bigger threat which, paradoxically, exists today years in advance of a powerful quantum computer existing. That threat is called harvest now, decrypt later (HNDL). If one assumes quantum computers will be built, it makes sense for an attacker to steal PKE-encrypted data today and decrypt it when one becomes available. Obviously, not all data is valuable but some such as military, government, financial and intellectual property data has a much longer shelf life.

There is no conclusive evidence that nation states or other attackers are stealing large amounts of data to conduct such an attack, but the assumption is that it is almost certainly happening in some form. This is no mere conjecture – senior U.S. Government officials have warned against this possibility.

Indeed, the involvement of governments in this field raises another alarming possibility. Perhaps a nation state will invent a quantum computer powerful enough to break PKE but not tell anyone. If so, this would represent the perfect hack, akin to the top secret breaking of German codes in WW2 by British mathematicians working at Bletchley Park.

Can encryption be saved?

NIST’s answer to the weakness of PKE is to adopt post-quantum cryptography (PQC) based on algorithms believed to be resistant to quantum computers, with the shortlist of candidates narrowed down to four in mid-2022. A second approach is quantum random number generators (QRNGs), which harnesses the same quantum principles as quantum computers themselves to generate encryption keys that are truly random (as opposed to non-random ones based on assumptions about difficult mathematical problems).

Quantum Key Distribution (QKD)

Generating secure keys is only half the problem – the keys must also be distributed securely enough that they can’t be quietly stolen by an eavesdropper. A technology which addresses this is Quantum Key Distribution. Based on quantum entanglement and using long symmetric keys (such as AES-256, which aren’t as easily undermined by quantum computers), this guarantees that if anyone eavesdropping on a key while it is being sent from A to B will find this impossible to hide according to physical laws. A QKD system responds to eavesdropping by sending a new key in place of one it knows has been intercepted.

The advantage of QKD over all the other quantum safe technologies is that it exists today as a first-generation technology and is even being used by UK financial sector companies.

Will quantum computers impact SMEs?

If you’re an SME, not for years. The organisations with the most to worry from harvest now, decrypt later attacks today are financial services, or those in the government or military supply chain. Nevertheless, the issues discussed in this article will gradually become more pressing over the next decade for everyone, mainly through having to migrate to post-quantum cryptography. Adopting PQC shouldn’t be hugely expensive but will still take up the time of network engineers and that implies extra work and costs.

What companies shouldn’t do is dismiss the issue as another scare story that will never come to pass. This is not like Y2K, which at least had a fixed date associated with its risks. Quantum computers could break PKE encryption at any moment, perhaps as early as 2030. When that happens, security will become critical overnight.

Conclusion:

Two things mark the quantum threat out as unlike any other problem in the history of computer security. First, it’s a threat we know about ahead of time. That is surely a good thing because it gives us fair warning. However, second, and strangest of all, it represents a threat in advance of its existence. This is the bit that’s hard to get to grips with – a future quantum computer could unlock secrets that are being sent today. What can we do about that? Probably very little. Nevertheless, this should be a spur. One day, security will have to be rebuilt to be ‘quantum safe’. That implies the alien but striking possibility that data security will no longer be a nice-to-have choice but the philosophical basis of absolutely everything.