Opinion

The future history of quantum computing

Nils Gerhardt, CTO at Utimaco, sets out the next stages of quantum chips
By
Nils Gerhardt
By
IBM Quantum Computer Mixing Chamber

For decades now, Quantum Computing has been a ‘when’, not an ‘if’.

The science for using the effects predicted by quantum mechanics has been speculated upon since 1959 before being formalised in 1985. A working quantum computer was first built in 1998, with only two qubits – far less than the 1000 or more that is the current state of the art, but it was able to carry out some basic calculations. The last 40 years have been spent overcoming the significant technical hurdles to creating a working quantum computer, a race for ‘quantum supremacy’ that has mostly involved adding ever more qubits – currently over 1,000 is the record.

Much has been made of the security implications of quantum computing: because they are so vastly more powerful than conventional computers, they can hypothetically break the encryption that underpins all of digital security. This won’t be something that happens overnight, but there will be a series of steps towards a post-quantum future that will need to be taken before we can see whether quantum computing will be a transformative new technology, a major threat or both.

In this article we will look at how the next decades of development are likely to play out. There are no guarantees with technology, but we can see from history and the current technological challenges faced by existing quantum computers what roadblocks may be coming up and how they may be overcome.

In the Next Few Years: Error Correction

The first and most pressing problem for quantum computers that needs to be solved in the near-term is error correction. To quote IBM, one of the few companies who could plausibly produce a quantum computer:

At IBM Quantum, we recognize that fault-tolerant universal quantum computers won’t roll off a manufacturing line tomorrow. Rather, we expect that quantum computers will mature incrementally as hardware and error-handling technology advances, steadily increasing in utility and solving increasingly complex problems in the interim, much like classical computers did before us.”

Currently, around one ‘bit flip error’ in which a zero accidentally becomes a one or vice-versa occurs in every hundred operations, which makes the current generation of quantum computers practically useless, especially at mathematically-intensive operations like breaking encryption. These errors would have to be reduced to one in a trillion for quantum computers to be as reliable as traditional computers, and as IBM notes, there is debate about whether physical error rates will ever fall below one in ten thousand. If this is the case then workarounds will need to be found- the simplest being running each operation multiple times and using the most common result, though this would mean having exponentially larger quantum computers.

Error correction would also mean reducing the external factors that cause noise in the first place, and this would be challenging. Often, the individual quantum bits that power quantum computers need to be kept at near absolute zero to avoid higher temperatures interfering with the single atoms that make up qubits – as anyone who’s taken high school-level science knows, atoms vibrate more at higher temperatures. These vibrations are the source of errors. Even interactions with outside atoms can disrupt the incredibly sensitive state of the qubits, so quantum computers need to be shielded or simple things like electromagnetic interference that a standard computer could shrug off could cause fatal errors. There are computers designed for operating in extreme environments like space which are hardened against interference, and their design might point the way to how future quantum computers will need to be shielded.

Solving these problems is going to be the most important technical hurdle in the near term for the developers of quantum computers.

In the long-term: Commercial Quantum Computing

Presuming that this problem can be solved (and there is every indication that it can be), how might quantum computers develop?

Clearly, a system with a massive number of qubits maintained at near absolute zero with electromagnetic shielding used in space stations isn’t going to be on anybody’s desktop soon, much less in their pockets. It is almost certain that the future of quantum computing will proceed much like the past of conventional computing, with quantum computers existing as massive mainframes in largely government, military and research roles.

Commercial applications are likely to be accessible through the cloud like Google’s TPUs, though they are likely to be very expensive to use at first, available only to large companies. This will mean that non-state hacking groups are unlikely to be able to access them for years or decades, or at very least find it difficult to do so. During this time, the only major threats from quantum computing are likely to be to and from state actors and companies, like defence contractors, who work with them. We have seen in previous conflicts that states will target utility companies or even private companies during conflicts, so it may not be the case that only government and military-adjacent companies need to worry.

This does not mean that the data your company needs to stay confidential today is at no threat. If, for example, a bad actor was to take a large amount of encrypted data from a company today they would be very unlikely to be able to decrypt it – breaking RSA-2048 encryption with a conventional computer would take approximately 19.8 quadrillion years. When they finally have access to reliable quantum computing, breaking that encryption could be trivial.

Preparing for the future today

The last point underlines why we need to start preparing for a post-quantum future today. We genuinely don’t know when commercially viable quantum computers are likely to arrive, especially since some developments are likely happening in secret. We do know that when they do some of the data that couldn’t be compromised today will be vulnerable. This wouldn’t just be twenty year old emails – it could be details of bank accounts that are still being used, patent protected data, sensitive health data, or, because one in ten people reuse passwords, passwords that are still being used. If state actors have access to this technology first, then they could falsify older data or create identities with ‘evidence’ stretching back years.

That’s why it’s paramount that companies working with any kind of customer data start looking at what they are storing today and what it could be worth tomorrow. Even though we don’t have anything like a timeline for the future of quantum computing, we do know that there are quantum-resistant forms of cryptography that can be implemented today, and that they can be implemented alongside the cryptography that you use now.

We don’t know exactly when and how quantum computing will arrive, but we can see from history its likely path and the current technical limitations that it has to overcome to get there.

For more information, visit: https://utimaco.com/

About the author

Nils Gerhardt has 19 years’ experience in the cyber security industry. In his current role, Nils is the Chief Technology Officer and head of product for Utimaco, a leading provider of cyber security solutions, and supervisory board member of ISITS AG. Before joining Utimaco, Nils worked at Giesecke + Devrient in various executive management roles with regional and global responsibilities in Germany, Canada, and the USA. As Chairman of the Board of GlobalPlatform, a global industry organization, Nils brought major companies together and led collaborative efforts to establish standards for secure global digital services and devices.

About UTIMACO

UTIMACO is a global platform provider of trusted Cybersecurity and Compliance solutions and services with headquarters in Aachen (Germany) and Campbell, CA (USA). UTIMACO develops on-premises and cloud-based hardware security modules, solutions for key management, data protection and identity management as well as data intelligence solutions for regulated critical infrastructures and Public Warning Systems. UTIMACO is one of the world's leading manufacturers in its key market segments.

550+ employees around the globe create innovative solutions and services to protect data, identities and communication networks with responsibility for global customers and citizens. Customers and partners in many different industries value the reliability and long-term investment security of UTIMACO’s high-security products and solutions. Find out more on www.utimaco.com.

Written by
Nils Gerhardt
CTO at Utimaco
April 24, 2024
Written by
April 24, 2024