Estimated reading time: 5 minutes
New standards for future proof encryption are close.
The potential for quantum computing to revolutionise work in sectors such as meteorology metallurgy and medical research by accelerating the processing speed of difficult calculations is relatively well known. So is the threat that quantum computing may render some current methods of cryptography insecure, weakening the security that protects how we work, shop, bank and live online.
But how far off are quantum computers really, and when should companies start to invest in “post-quantum security”?
The answer to both questions is “sooner than you think”.
New standards for cryptographic algorithms that can protect data against quantum attack are close to being finalised.
“The Federal Office for Information Security in Germany has already published some of its evaluation and recommendations for post-quantum cryptographic algorithms,” says Thales’ Director and Senior Advanced Cryptography expert, Aline Gouget.
In the US, Gouget continues, standards setting body the National Institute of Standards and Technology (NIST) is in the final stages of evaluating seven final candidates and eight alternatives for post-quantum cryptography.
They may not hit mainstream news often, but major developments in the world of quantum computing are happening all the time. In this first few weeks of 2021 alone, one group of Chinese scientists revealed the world’s first quantum communication network, while another Chinese team launched the county’s first home-grown quantum operating system. The French government, meanwhile, announced a €1.8 billion plan to invest in quantum computers and related technologies, and IBM updated its roadmap for its quantum computing development confirming that it aims to have a 1121 qubit processor in operation by 2023.
What is post-quantum security?
One of the functions that it is believed quantum computers will be much, much better than conventional processors at is cracking “public key cryptography”. Public key cryptography is used for securing everything from personal emails, to financial data when you log into a banking app, to the instructions sent to an Internet of Things device.
Because current methods of public key cryptography are vulnerable to attack by quantum computers, researchers around the world are developing new cryptographic algorithms for post-quantum security.
Post-quantum security does not require quantum computers
It’s important to note that post-quantum cryptographic algorithms do not require quantum computers to create or decrypt information between authorised parties.
• They protect “brute force” attacks using quantum computers against encrypted data.
• Not all current cryptography is vulnerable to attack using quantum computers. Symmetric cryptography, such as the AES security commonly used to encrypt files at rest, is not known to be at risk.
What is public key cryptography?
Public key encryption is a very common form of cryptography used to secure communications.
It uses the maths of prime numbers to encrypt messages using a key that the intended recipient of the message has shared with the person sending the message. Only the intended recipient, however, has the private key that can decrypt the message.
The reason public key encryption is ubiquitous is that you can share your public key by publishing it for anyone to access, safe in the knowledge that it cannot be used to decrypt messages sent to you. It makes sending encrypted messages very easy.
Breaking this encryption without the private key would mean finding the “prime factors” used to create the public key. These are two prime numbers which are multiplied together as part of the encryption process to form part of the public key.
Because data is encrypted with the public key but decrypted with the private key, it is a form of “asymmetric cryptography”.
For sufficiently large prime numbers this is considered an impossible task for today’s computers.
In theory, however, quantum computers should be good at prime factorisation and therefore able to decrypt messages using only the public, and not the private, key. Mathematics that would take thousands of years on today’s technology could be reduced to hours on a quantum machine – and much of today’s security would be obsolete.
So designing security for the post-quantum world, or “post-quantum security” (also known as “post-quantum cryptography”) means new techniques and algorithms must be adopted, standardised and widely used.
Although the development of quantum computers poses a challenge to current security, it does promise many benefits which outweigh the risks.
• Accelerometers and navigation systems - “Quantum sensors can improve the accuracy of GPS systems by a factor of 100, or maybe more,” says Marko Erman, Chief Scientific Officer at Thales.
• Quantum encryption - “New encryption techniques which make use of the quantum properties of light particles over fibre optic cables could improve encryption techniques even further”, says Erman.
• Drug design and chemistry - One of the key promises of quantum computers is their expected ability to model complex systems in more detail than current computers. The application of this could revolutionise drug design and chemistry.
• Quantum sensors - Quantum technology can vastly improve antennae, radar and electronic warfare systems,” says Erman. “The prototypes we are developing significantly outperform conventional systems, they offer superior detection capability across a broad range of frequencies.” In addition, quantum-based sensors can be much smaller than traditional base stations – from several square meters to palm-sized devices.
The roadmap to post-quantum security
“The move to public key encryption as a standard was very challenging,” says Gouget.
Current public key encryption has been in use for three decades but it took a long time to become standardised. There have been lessons learned from that process and early moves to standardise post-quantum are promising, but challenges remain.
“What we do with current cryptography is reuse the public key lots of times, but with some methods of future key encapsulation this is not possible.”
Businesses should not wait to begin preparing their own roadmaps, though.
The first thing to do, says Gouget, “is to take an inventory of what cryptography you are using and how long the data it is protecting must be secured. If it is 30 years and the algorithm is only safe for ten years, you have a problem you must plan to address”.
In other words, now is the time to prepare for the quantum future, because it’s closer than you think.
The positive side-benefits of post-quantum cryptography
Gouget believes that the final standards for post-quantum security will likely involve “some combination of current cryptography and a hybrid that is safe from quantum attack. Many of the quantum safe algorithms currently being investigated used lattice-based cryptography, and Gouget says that expertise being developed in this branch of mathematics may have other benefits too.
It may speed the development of robust “homomorphic cryptography”, for example.
Homomorphic cryptography is an emerging technique which enables datasets to be processed in an encrypted form. In other words, it will enable one organisation to share data which might contain sensitive information with another organisation that can process it without every seeing it in unencrypted form. It has strong potential applications for protecting personal privacy, for example, while still enabling big data processing.
In quantum computing, a “qubit” is comparable to the “bit” in traditional computing, in that it is the smallest block of information that a quantum computer can operate on.
• A qubit is a sub-atomic particle
• Information is stored and read from a qubit using quantum mechanics
• While a desktop CPU can only perform one operation on a bit at a time, a quantum computer can perform multiple operations on a qubit simultaneously.
• Adding another qubit increases the number of operations that can be performed exponentially.
Writing for IEEE Spectrum magazine, Charles Q Choi says that “In principle, a quantum computer with 300 qubits could perform more calculations in an instant than there are atoms in the visible universe.”
There are many ways to build a quantum computer, and the complexity of building and programming a quantum computer means that we are still some way from realising this kind of computing power. But if IBM achieves its aims, says Gouget, “Then after that there will be few blocking points, because they will be able to scale”.