Seeking Security in a Post-Quantum World

Government agencies are rapidly moving to counter the threat posed to secure encrypted messaging by quantum computing

For 30 years, researchers have known that quantum computers, when they become reality, will compromise the security of all electronic communications, rendering widely-used existing security protocols, like those used in financial messaging services, useless.

Government agencies, including the security services and national standards bodies, the academic research community and developers at tech companies like IBM, Google, Amazon, Microsoft and Intel are all now accelerating their responses to the potential threat.

Current communication protocols are based on agreed standards such as public key infrastructure (PKI) and variations like the RSA cryptosystem. The algorithms that underlie these protocols – which have taken more than 20 years to develop and deploy – are based on the factorisation of very large prime numbers, putting the mathematical complexity of solving them beyond the capability of even the most powerful supercomputers using binary logic.

Quantum computing is now approaching reality and, because it is not restricted to binary logic, it will be able to solve these problems much, much, faster than conventional machines. It would take a classical supercomputer billions of years to break an RSA-encrypted communication using brute force: according to one estimate a quantum computer with 8,000 qubits (the measurement of compute power in quantum computers) could solve the problem in just eight hours using an algorithm that already exists.

There are other factors that need to be considered other than just qubit counts: most importantly, current quantum devices suffer from instability and very high error rates, so a linear growth in compute power along the lines of Moore’s Law for semiconductors isn’t applicable. Other estimates suggest that the redundancy required to overcome these flaws would require 20,000 qubits. But even so, the current rate of development suggests that the capability to ‘crack’ a message encoded using the RSA 2048 standard (which requires a 2048-bit public key) will be achieved sometime between 2030 and 2035.

The fastest quantum device currently in development is IBM’s Condor chip, announced at the end of 2023, which is rated at 1,000 qubits. IBM hopes to reach 4,000 qubits sometime in 2025. Although it might seem that the point at which the power of quantum computing is commercially available at a large scale is a way off, IBM’s devices had five qubits just eight years ago. As such 2030-35 is likely a reasonable estimate.

The sort-of good news is that almost all of the development of powerful quantum devices is being carried out openly by the likes of IBM, Google, Amazon, Intel and Microsoft, working with government and academic labs .

This is largely because of the cost of developing quantum technology, which puts it into a bracket that until recently was the province of nation states, and companies that have budgets larger than most nation states. It is not known what proportion of IBM’s R&D budget – $6.78 billion in 2023 and $3.6 billion in the first half of 2024 – is allocated to quantum computing, but it is likely to be substantial. It has also committed $100 million to academic research in the US and Japan (topped up by a further $50 million from Google). The field is also not short of private money funding start-ups either: according to McKinsey, quantum technology start-ups raised $2.35 billion in 2022.

These sums are insignificant compared to government investment, however. McKinsey points to the $1.8 billion the US has committed to its National Quantum Initiative Act, and the £7.2bn spent by EU member states on quantum research programmes.

These, in turn, are dwarfed by China’s $15.3 billion investments, putting it well out in front in terms of spending – which is backed up by considerable success in quantum breakthroughs, including a satellite communication system using quantum entanglement, a phenomenon in which pairs of subatomic particles can be made to influence each other over great distances without any connection.

While these investments represent another facet of the country’s drive to overtake the US and Western economies in advanced technologies, it is also suspected that they are being used for cyber-espionage and related activities.

While China is, to some extent, a “known known” in terms of its capabilities and ambitions, other nation states and their proxies are actively hostile. These pose a hazier threat but as these states often outsource to organised crime, they are already one of the most prevalent attack vectors faced by financial institutions of all kinds. What quantum cryptography capabilities would add to their toolkit are two new entry points: the first is the “harvest now, decrypt later” attack, where encrypted data may have already been stolen and saved for later decryption; the second is the threat that it would pose to blockchain or digital ledger-based financial networks.

The first stage in countering these threats is to build higher walls and better locks, which is what is currently underway, most prominently in the US.

In May 2021 an Executive Order signed by President Biden gave several US government agencies the task of reviewing and improving the country’s cybersecurity defences across the board, including multi-factor authentication, the adoption of Zero Trust Architectures when migrating systems to cloud platforms, data encryption and pretty much anything else they could think of.

At the forefront of this defensive effort is the National Institute of Standards and Technology (NIST), which was set up in 1901 with the intention of boosting the US industrial infrastructure – then lagging the UK and Germany – and to create national standards to increase industrial efficiency (which sounds like a similar initiative to the current Chinese technology effort).

In the past few years, it has developed the NIST Cybersecurity Framework (CSF) to provide guidance “to industry, government agencies, and other organisations” to manage cybersecurity risks. Its taxonomy of high-level cybersecurity outcomes “can be used by any organisation,” although if the organisation in question does any business with a government body, they are obliged to adhere to the CSF.

Last month, CSF 2.0 was upgraded to include three new algorithms “designed to resist future attacks by quantum computers, which threaten the security of current standards”.

These standards specify key establishment and digital signature schemes using algorithms that are derived from different submissions in the NIST’s post-quantum cryptography standardisation project.

Similar initiatives are being followed in most other countries. In the UK, the National Cyber Security Centre (NCSC), part of the Government Communications Headquarters (GCHQ) provides post-quantum cryptography guidance to organisations and advises on preparing for a transition to quantum-safe algorithms. The NCSC is working in partnership with the European Telecommunications Standards Institute (ETSI) and the International Organisation for Standardisation (ISO) and others on international efforts to develop and standardise post-quantum cryptographic algorithms.

The Monetary Authority of Singapore (MAS) is also moving quickly: following an advisory note issued to banks and other financial institutions in February 2024, MAS has assembled a consortium of banks – including DBS, HSBC, Oversea-Chinese Banking Corporation (OCBC) and United Overseas Bank – and tech companies to pilot the use of post-quantum cryptography and Quantum Key Distribution (QKD) to safeguard critical data.

It has often been observed that among the early adopters of new technologies are those who see the possibilities of using them for nefarious purposes: the start of the quantum computing age is unlikely to be any different. To reap the benefits of its potential will require strong safeguards. Starting now.

---