Emerging Computing Paradigms and Future Security Threats

This publication is licensed under the terms of the Creative Commons Attribution License 4.0 which permits unrestricted use, provided the original author and source are credited.

Introduction

Digital computing has been the dominant technological paradigm for decades, but researchers now have a growing interest in alternative computational systems. These systems include analogue mechanical and electrical computational systems, light-based computing and – most notably – quantum computing. Such systems are usually designed for specific applications, purportedly solving particular computational tasks more efficiently than digital computers can.

In an adversarial setting, policymakers should not view potential attacks using most new computational systems as a pressing concern. The key exception is quantum computing, which presents a substantially greater challenge than the others. This is partly due to a broad programme of research to develop quantum technologies, which is advancing at a greater rate than those for other approaches to computing.

For example, the United Kingdom is continuing to build upon its history as a leader in the space with the National Quantum Technologies Programme. This follows strategic investments in ancillary areas such as cryogenic cooling systems, which have given the country a significant advantage. Last year, the UK released the National Quantum Strategy Missions, setting out a series of development goals in key areas of quantum technology, not least quantum computing, networks and sensing.

The upshot is that UK regulation will need to sustain the current pace of research while mitigating the threat of quantum attacks. This analysis will examine several quantum-based threats, starting with a brief overview of those in post-quantum cryptography. It will then survey the legislative restrictions on the export of quantum technologies that many countries have introduced in recent years. Finally, it will make policy recommendations for the UK government in responding to these risks and opportunities.

Post-quantum cryptography

Since the development of Shor’s algorithm in the 1990s, it has been clear that quantum computing poses a threat to cryptographic systems. A functioning quantum computer could break into any system based on the RSA standard or the discrete logarithm problem. Together, these make up most current encryption and digital signature schemes. Concern over this potential threat prompted the US National Institute of Standards and Technology (NIST) to lead a re-standardisation process to replace existing cryptographic protocols with quantum-secure methods.

Current quantum computers are only able to factor 21 into 3 and 7 – a far cry from the 1,000+ digit factorisation required to be an active threat. However, there are more immediate threat vectors in the form of the ‘store now, decrypt later’ method, which involves the interception and storage of large amounts of encrypted data. Once a functioning quantum computer becomes accessible, this data could be decrypted and exploited for nefarious means. For instance, the architectural plans of secure buildings – such as military bases – may still be useful to adversaries on a time scale on which one could expect quantum technologies to be developed. As of August 2024, the NIST had released three post-quantum encryption standards, with a fourth to follow later in the year. The organisation is also evaluating other backup standards.

Quantum side-channel attacks

Threats to information security fall loosely into one of two categories. The first includes direct attempts to break protocols. These are the quantum attacks that have necessitated new post-quantum cryptography standards.

The second category includes methods that exploit additional information on the adoption of these protocols, such as that on discrepancies in their implementation and on the hardware they use. Such methods can lead to a security breach by revealing passwords, encryption keys and more fundamental information required to break protocols. Threats in the second category are known as side-channel attacks (SCAs). They encompass several key methods for extracting information required to break cryptographic protocols:

• Timing attacks gather information on discrepancies in the running times of protocols, particularly the algorithms they use.
• Energy attacks measure the energy usage of a computer running a cryptographic protocol to identify inconsistencies.
• Audio attacks, which are less common, measure noises emitted by devices.

It is important to carefully consider SCAs when designing and implementing communications security systems, given that such attacks have been the cause of many leaks and security breakdowns.

One burgeoning research programme centres on the use of a range of quantum technologies in side-channel attacks (SCA-QS) and methods to defend against them. The focus here is on quantum sensors rather than quantum computation. These instruments use quantum mechanical properties to enhance classical methods of measurement, often markedly outperforming their established counterparts in accuracy and scope of measurable phenomena. They include well-known devices such as atomic clocks and electron tunnelling microscopes, as well as instruments for system positioning, communication, seismology and electromagnetic-field sensing.

Three of the five UK National Quantum Strategy Missions involve research into quantum sensors and their use in a plethora of settings. Until the advent of the SCA-QS programme, SCAs and quantum sensing fell into disparate fields of research. However, there are some clear use cases for quantum sensors in SCAs. For example, in 2007, the Xbox 360’s security systems were broken using timing attacks. An algorithm that formed a crucial step in the security protocol ran at different speeds depending on the input it received, allowing an attacker to fake an input to breach the device’s security.

While Microsoft addressed the vulnerability by ensuring that the algorithm in question ran at the same speed regardless of input, such solutions may only be effective in a classical setting. If an attacker used a quantum sensor to identify inconsistencies that were otherwise undetectable, this could present a major security risk. Public research into such threats from quantum technology is still nascent, even though quantum sensors have been highly functional for a long time.

Threats to quantum communications networks

Beyond concerns about quantum technology threatening classical security systems, it is important to consider the security of quantum communications networks. Recent decades have seen the emergence of proposals for networks that connect quantum computation systems by sending and receiving quantum information exclusively. In this case, quantum information refers to quantum bits or qubits: the quantum counterpart of digital bits and binary strings, containing the kind of richer information that allows quantum computation to outperform classical computation. There are now several small-scale quantum communications networks in place, with larger ones under consideration.

Currently, communication between quantum computers relies on the translation of quantum information into the more familiar digital information that classical computers use, transmitting this and inputting it into another quantum computer. Quantum networks, by contrast, can set up a direct link between quantum computers, removing the need for translation into classical information. These networks can take advantage of already existing fibre optic infrastructure to transmit quantum information, but may also require the construction of new infrastructure. Like digital communications networks, quantum networks come with their own set of security concerns and attack models.

Work on quantum cryptography began in the 1980s but, for much of this time, it has been confined to the theoretical realm. Given the networks already in place – and the drive to develop “the world’s most advanced quantum network at scale” in the second of the UK’s National Quantum Strategy Missions – there is a need to explore, and develop regulatory requirements for, existing protocols.

How advanced is quantum technology?

There is a wide variation in predictions of when quantum computational technology will become a credible threat. With the post-quantum cryptography re-standardisation procedure, the NIST is taking worthwhile precautions in the development of this technology and simultaneously guarding against ‘store now, decrypt later’ attacks.

The UK’s National Quantum Strategy Missions set out a goal that, if it were achieved, would lead to the development by 2035 of a quantum computer capable of launching attacks that threaten classical cryptography (even though this is not the driving force of the missions, and that timeline may be unrealistic). Greater uncertainty surrounds the other forms of attack discussed above. For instance, it is unclear how quantum sensors will be used in future SCAs. Therefore, rather than attempt to predict how future research will develop, it would be more prudent to produce a range of considered responses to advances in these technologies.

Legislative restrictions

The UK is one of a group of countries that have passed in recent years legislation restricting the export of quantum computers. Its measures came into force on 1 April 2024 as an extension of The Export Control Order 2008. Such restrictions are not a universal ban on exports of the technology but impose limits on the size and functional capacities of applicable devices.

Countries including France, Spain and Canada have implemented the exact same limits. The reasoning behind the restrictions is unclear, with several reports simply stating that they were chosen ‘based on scientific analyses of quantum computers’ and applied to technology that was, on some level, ‘likely to represent a cyber risk’. However, each country that has introduced this legislation is one of 42 signatories to the Wassenaar Arrangement, an export control system for technologies that could have military applications.

While this indicates that there are national security grounds for the legislation, it also appears to have a secondary economic drive. The industry that produces semiconductors, the main component of chips in digital computers, underwent rapid growth in America and Asia in the late twentieth century. The process left European countries on the back foot despite their promising start in the area. Some of them, including the UK, are currently at the forefront of quantum technological research and want to avoid repeating the mistakes of the past. However, these restrictions have met with resistance from academics and private companies, which warn that they will severely limit research progress and innovation.

This is an evolving interest for the UK, which has positioned itself to become a leader in setting global standards for quantum technologies. As part of the effort, the country has established and led the first international quantum technology committee, focusing on the development of best practices and standards. The committee’s work should result in more nuanced regulatory and legislative responses to quantum technologies, ideally finding the right balance between technological development and both national and economic security.

Policy recommendations

There is already a clear policy path for the re-standardisation of post-quantum cryptography. The National Cyber Security Centre (NCSC) initially published a white paper recommending that the UK wait for the NIST process to conclude before responding to the resulting standards and guidance. Now that the NIST standards have been released, UK policy should integrate and adopt them, as reflected in the updated recommendations from the NCSC.

The NIST states that ‘system administrators’ should start integrating the standards ‘into their systems immediately, because integration will take time.’ Yet, despite the pressure to do so quickly, it is vital to conduct the process in a manner that minimises security risks. It is also important for the government to work with industry leaders to ensure widespread adoption of the standards, many of which will be applicable to technology used in everyday life. Policy and regulatory decisions should reflect this, working to a timeline suited to the complexity of the task.

Policy on the other quantum technologies discussed above will necessarily be more nuanced and disparate, due to their experimental nature. With these technologies at varying stages of maturity, and with no clear picture of their development timelines, there need to be case-by-case assessments of their security implications for classical technologies that are susceptible to quantum attacks. While the uncertainty that surrounds them complicates advance planning, there are ways to encourage positive technological innovation and flexibly respond to the emergence of new security threats.

One solution could be to propose a set of technical milestones for quantum technologies, each with a related policy response. This approach already features in regulatory discussions of quantum technology enterprise put forward by the Regulatory Horizons Council, which the government accepted in a recent policy review. Such responses are based on technology readiness levels (TRLs), a general framework for characterising technological advances. When the UK authorities consider a quantum technology to have reached a particular TRL, they can turn to a predefined regulatory response. The UK should develop a similar approach to the security aspects of quantum technology. It will be vital to understand the full range of policy implications of this technology as soon as possible rather than react to unfamiliar threat horizons.

Finally, while legislative restrictions on the export and sharing of quantum technologies are important to national security, they will slow technological development in the UK. The creation of sharing agreements with strategic partner countries would both accelerate the development of beneficial technologies and help the UK recognise when it had reached a milestone in the area. 

The views expressed in this article are those of the author, and do not necessarily represent the views of The Alan Turing Institute or any other organisation.