As computer scientists march forward in the process of bringing quantum computing into the practical world, cybersecurity vendors and practitioners will need to be ready with encryption mechanisms that can withstand the power of quantum computing potential. But risk experts say future-proofing measures for post-quantum cryptography need not be created in a panic.
Contrary to the way some early experts have painted the post-quantum computing landscape, the truth is that there will be no quantum tipping point where today’s encryption mechanisms suddenly become obsolete, says Dr. Colin Soutar, the US head of quantum cyber preparedness and CEO of Deloitte Risk & Financial Advisory, who just released a report on quantum cryptography. He explains that the transition to quantum will in fact be an ongoing process.
“There’s a lot of discussion around quantum right now, and there’s a lot of conflation of different ideas. There are even some alarming statements about how everything has to change overnight to update to quantum-resistant algorithms,” says Soutar . “It means there’s a set date (for quantum adoption), and there really isn’t.”
Viewing post-quantum security issues from that kind of lens can help the cybersecurity industry begin to address the issue with the same kind of risk management and planning steps that they would take for any other kind of serious new technology trend.
Awareness building, not alarmism
One thing is certain: the drumbeat for quantum computing and post-quantum cryptography is getting louder.
Quantum computing stands to give the computing world a big boost in the ability to tackle multidimensional analysis problems that burden today’s most advanced traditional supercomputers. While traditional computers basically work based on storing information in binary, quantum computing is not limited by the “on” or “off” position of information storage.
Quantum computers rely on the phenomenon of quantum mechanics called superposition, where a particle can exist in two different states simultaneously. They exploit this phenomenon by using “qubits”, which can store information in a number of different states at the same time.
When perfected, this will give quantum computers the ability to greatly speed up data analysis of hard problems in fields as diverse as health research and artificial intelligence. But this kind of power also makes these computers ideal for cracking cryptographic algorithms. This is at the heart of the push for awareness by security advocates over the past several years to ensure that the industry begins to prepare for the post-quantum reality.
“Our view on this is less about being alarmist and saying, ‘You need to update everything now,’ and more about raising awareness to start thinking about what your data is, what your risk might be in relation to that data and the crypto you use,” says Soutar. “And then decide when you might want to reconsider, start looking at discovery on your roadmap, and then update later.”
According to the survey released by Deloitte this week, the good news is that among the technology and business leaders aware of quantum computing, just over 50% also understood the associated security considerations.
Timing of the post-quantum security impact
The trick in all of this for security professionals is that there are a lot of fires to put out elsewhere before you worry about something that could be years away. Today’s quantum computers operate only in the research area. They require enormously specialized equipment—including microwaves that manipulate quantum objects in supercooled environments that operate at near absolute zero in many cases. There is a long way to go on the research front for quantum computers to work in a commercially viable way, and no one is quite sure what the timeline will be.
That “timeline ambiguity” is complicated, says Soutar, who explains that there are several timelines to consider from a post-quantum cryptography perspective.
“The implications of quantum computing on cybersecurity are pretty well known, and it could be huge. I mean, cryptography is endemic in what we do throughout the economy. The thing is, the timing is unknown because first a quantum computer has to be mature and viable enough and also commercially robust to actually be able to run Shor’s algorithm,” he says, referring to an algorithm for finding prime factors of an integer that is the benchmark for whether a quantum computer could effectively break public key cryptography. “Second, attackers need to get access to data and they need to solve that data.”
The other variable in this is an attack concept called “harvest now, decrypt later,” where attackers collect encrypted information now with the understanding that they could break it through quantum computing resources at a later time. The Deloitte survey shows that 50.2% of organizations believe they may be at risk of harvest now, decrypt later schemes.
“It opens up the risk of this data, which I expect will be good for a person’s lifetime,” says Soutar. “Maybe it’s personal information, or it’s financial information that I want to be secure for at least 10 years. Or it’s national security information that may have longer claims.”
He adds, “So people are starting to think, ‘Well, what data do I have and how do I protect it?’ How long? Second, how long will it take me to make the updates to send them quantum cryptography? When do I start thinking about it?'”
These are the big timeline questions for security and quantum computing experts, who still disagree about whether we have 5, 10 or 15 years before the quantum effect affects encryption. Soutar reiterates that perhaps the better thought process is to stop thinking of it as a definitive date the industry takes time to, and instead think of relative risk over time. He explains that this is an idea put forward by Dr. Michele Mosca, co-founder and CEO of Evolution Inc, and co-author of a report earlier this year detailing that thinking.
“Then you can start thinking that if I’m with a huge organization, it might take me a decade to do the updates,” explains Soutar. “I have all this medical equipment or other OT equipment that I have to think about supply chain communications, and how do I enforce this with my suppliers?”
He adds: “Then again, it gets the right level of understanding so that people can start to maybe even quantify what the risk is and stack it up against other cyber risks that they’re looking to invest in over time.”
Working on the boring parts
At the end of the day, Soutar says the quantum lens might be a bit of a distraction for security. As long as organizations keep quant on the horizon, it may just be a matter of making “shallow updates to crypto,” which may not be such a big deal for the industry if it’s all done in time.
“The quantum threat to crypto should really just be something that gets addressed over time. Just make updates as the algorithms become standardized,” says Soutar, who believes the industry should be talking about the nuts and bolts of standardization, which can be boring but is also the most important way to start moving forward. “Once they go through that process, then companies and governments have more confidence to make the changes, make the updates, and they just do it. So it really should be a non-event.”
That’s not to say Soutar thinks security practitioners should bury their heads in the sand regarding quantum risk for security positions. The risks will accelerate, but it’s just a matter of working with that encryption roadmap like any other part of the cyber risk roadmap. It includes making risk assessments, discovering and classifying data and projecting risk over time.
“It’s never a bad idea to go and look around the attic. You never know what you’ll find there. When we do that, when we go through basic cryptography, there are things we find,” he says. “You can say, ‘Well, let’s update it, or let’s make sure we have the proper segregation of duties on that.’ Or, ‘Have we laid out all the responsibilities and governance?’ Again, it’s the boring stuff. But it’s stuff you find when you look through the quantum lens.”
Deloitte’s study shows that it may take some kind of regulatory push to get security practitioners to take serious steps with post-quantum cryptography. Soutar hopes that the industry is able to come together in the coming years to develop a framework for post-quantum cryptographic methods perhaps in the same spirit as the NIST Cybersecurity Framework (CSF).
“It’s not a bad idea to have some frameworks out there when there’s a hint of potential regulation downstream,” he says. “I think it’s always better than just regulation to have something that’s voluntary and results-based.”