In 2001, NIST (the US National Institute of Standards and Technology) announced Advanced Encryption Standard (AES), a new encryption standard, designed to help organisations enhance protections against brute force attacks. The previous Data Encryption Standard (DES) had become vulnerable, with processing power growing, and the Electronic Frontier Foundation (EFF) had proved that DES encryption could be broken in less than 24 hours, therefore a new encryption standard was required.
That was more than two decades ago, and AES has served us well, but are its days numbered?
I am not the only one asking this question. NIST has yet again called for new encryption standards, this time recognising that quantum computing will eventually become a threat to our current encryption standards. The consultation is already well underway, with the four finalist contenders for Quantum-Resistant Cryptographic Algorithms announced.
The reason for all this activity is simple—quantum computers can use their advanced mechanical processing advantage to crack encryption that previously took an unworkably long time to break. And while today’s quantum computers are currently found in laboratory and research settings, there is talk that in 2023 we could see the first quantum computers go on general sale. Once these machines get into the wrong hands, our aging encryption methods could quickly become compromised in a focused quantum decryption effort.
In reality of course, these resources don’t even have to be available on the open market to be a threat. State-sponsored actors are behind the majority of large and successful cyber attacks today. Given the funding many states put towards cyber warfare and cyber espionage, it is not far-fetched to imagine that they will be early adopters or even look to gain access to the quantum computers that reside in universities and other research institutions today.
And perhaps quantum only needs to be a possibility to spur a data breach. The phrase “hack now, crack later” describes cyber criminals anticipating future decryption capabilities and stealing data that they cannot currently decrypt and access today. Think of it as a stolen nest egg.
In this blog headline I referenced a quantum arms race. Many states have publicly declared their interest in getting to quantum first, or at least in the vanguard. It is a widely acknowledged national security advantage as much as it is an economically driven aspiration. You don’t need to have watched all of the James Bond movies to worry about what might happen if the bad guys get to quantum before the good guys.
Given this context, it is inevitable that the security professionals I have been speaking with are starting to plan; working out the extent of change needed and beginning to anticipate impending disruption. As an outline, consider the following as to what organisations should start to think about and what may be helpful in a long-term encryption strategy. Think of it as a Solace for Quantum if you will…
Organisations should start to plan a process for an encryption standards switchover. This plan will need to factor in all platforms and products that are already in the security stack and network architecture, so teams should open channels of communication with all technology vendors to start discussions around how they are planning to tackle the encryption migration.
Bear in mind that applying quantum-resistant encryption—at least in its early years—is simply not going to be a workable approach for the bulk of your data. It could create huge performance issues if you are expecting to encrypt data that is regularly accessed and used (quantum computers are powerful, but it will take a while before the processing power is ubiquitous enough to be found handling the inline, dynamic decryption that we are all used to for transactions). More appropriately, now is a good time to consider zero trust principles and modern security architectures which will provide a solid foundation upon which to build encryption changes in time.
This is also yet another opportunity for me to advise undertaking a data audit—regular readers will know I see this as an essential starting point for any effective data protection strategy. Identify what data your organisation holds, where it resides, and how access is granted.
In the immediate period, I would recommend that you pay special attention to data which has long term value. For example, if someone stole it today, would it be worth sitting on it for 1, 3, 5, or 10 years or more until a simple decryption option is available to them? If the answer is yes, this is the data that you should prioritise as the new standards become available. Realistically, most of the data you need to protect today might never need to be encrypted using quantum-resistant methods—it will lose most of its value before quantum attacks are a reality, so your plans for re-encrypting run-rate data can be phased in for new data rather than retro-fitted against old.
I often get asked what a realistic timeline looks like for quantum security considerations. While we can only wait to see how quickly these machines come to the market, government deadlines provide a useful indication of the expectations of those in the know. As an example, the US Government has specified that all US Government agencies and their suppliers must use quantum-resistant encryption by 2035 (with roll-out set to start in 2025). That may seem a long way off, but it isn’t, and it’s even less time if,for instance, you are an organisation that is part of the supply chain to government agencies across the world.
Also, don’t simply look forward in this planning, you need to look backwards too. While I have said much of your current data may lose value very quickly, your organisation probably has a store of archived data. Don’t forget to evaluate the data protection you have surrounding this long-term storage of data. Identify how long you have held data and audit how it is protected. You may even discover you are still using DES or 3DES!
Finally, think about how you will explain these initiatives, standards changes and your organisation’s change requirements to all necessary stakeholders. All of this planning and activity is likely to impact both resource bandwidth and budget over the coming years. It will also be important to keep in touch with regulatory bodies—both on a national and industry level. Close collaboration will ensure you keep ahead of impending compliance issues and these bodies will also likely start to issue helpful guidelines over the coming months and years.
We can probably all agree that this kind of innovation is one of the most appealing characteristics of the industry for many of us. But any change requires thought and effort to stay ahead. I am looking forward to collaborating with my peers and supporting our customers as we move into this new challenge. Because, as Commander Bond might state, you always want to ensure sensitive data remains For Your Eyes Only.