Quantum computing: the dark side of computing power

Cryptosystems such as RSA are standard in today’s digital world. To date, these methods have relied on the assumption that certain complex mathematical operations cannot be reversed in realistic timeframes — a classic example being the factorization of large numbers into their prime factors. With the advent of quantum computers, however, this assumption could become obsolete. Algorithms specifically designed for qubits will very likely soon be able to create “shortcuts” for solving such problems — in other words, to break encryption. The day this first becomes practically possible is known as Q-Day. Google, for instance, estimates it could arrive as early as 2029.

Companies looking to protect themselves against the looming quantum threat should first gain as clear a picture as possible of their own exposure and then evaluate concrete technical measures on that basis.

The following steps can serve as a guide:

1. Risk assessment

Even though virtually every company operating in the digital space works with encryption in some form, not all information is equally at risk. Once quantum technology moves into practical use, it can be assumed that this will initially occur within a very exclusive circle — reserved for government bodies or major technology corporations. As technology develops further, however, it is reasonable to expect that quantum computing capacity will become more broadly accessible through Infrastructure-as-a-Service offerings – potentially used against high-value targets with criminal intent.

Companies should therefore start by examining what role digital intellectual property plays for them, in what form it exists, and whether such sensitive information is transmitted via encrypted channels.

2. Mapping encryption points

To reliably protect their infrastructure, companies must first build a detailed picture of their current state, including the various dependencies involved. In a world of hybrid and multi-cloud environments, this is less straightforward than it may initially sound. It requires being able to precisely trace the path data takes, knowing where and with which methods it is encrypted, and who controls the corresponding keys.

3. Planning crypto-agility

Building on their own risk assessment and an inventory of their various crypto assets, companies can determine where action is most urgently needed and which quantum-safe encryption methods should be introduced first. They should also check whether their cloud partners allow the use of their own encryption keys. If so, companies can retain full control and decide for themselves when the time is right to make the switch.

High-risk industries are advised to adopt next-generation algorithms already today, to guard against so-called “harvest now, decrypt later” attacks — in which valuable data that is currently protected by encryption is collected with the intention of decrypting it later using quantum computing power. The US National Institute of Standards and Technology (NIST) has already published corresponding standards for quantum-safe encryption.

This is by no means a purely theoretical discussion any longer. Both the US and the EU already have concrete timelines and deadlines in place for the deployment of quantum-safe algorithms. In the United States, critical systems must transition to post-quantum cryptography by the end of 2030, while the EU roadmap calls for high-risk applications to likewise be migrated by 2030, with initial implementation measures to begin as early as 2026.

4. Checking your own or partner hardware

Hardware security modules (HSMs) are the ideal choice for generating and storing cryptographic keys. Unlike software-based solutions, they can guarantee tamper resistance, making them suitable for critical applications as well. Depending on their operating model, companies should check whether their own hardware, or HSMs provided by partners, are sufficiently capable. HSM-as-a-Service offerings can today also achieve high levels of protection in line with standards such as FIPS or PCI.