We have a habit of turning things into slogans. Artificial intelligence becomes “the end of work.” Ransomware becomes “cyber warfare.” Quantum computing becomes “the death of encryption.”
Slogans are not always wrong, but they’re rarely nuanced. It’s impossible to know what is real, theoretical, urgent, just interesting, or what you should actually do about an exposure based on a slogan.
Quantum computing and encryption obsolescence is a prime example. There is a real exposure that deserves attention, planning, and preparation, but it doesn’t justify panic. It doesn’t mean all cybersecurity controls are obsolete or you need to rip out your tech stack tomorrow.
The sober view of the impending Q-Day is that some cryptography is expected to become vulnerable to powerful quantum computers. Those computers don’t exist yet at the scale needed to break modern encryption, but the migration away from vulnerable cryptography will be slow and complex. For sensitive data that needs to remain confidential for years, the exposure may already matter. We could be decades out, but it’s a risk worth understanding today.
Encryption
Encryption turns readable information into unreadable information unless someone has the right key.
When you log into your bank, encryption helps protect the information moving between your browser and the bank’s website. When a hospital stores patient records, encryption can help protect those records if a database is stolen by a cyber criminal.
Encryption protects two main things:
Data in transit. That means data moving from one place to another i.e. a browser to a website, an employee laptop to a VPN, an API request between two softwares, or sending an encrypted text message.
Data at rest. That means data sitting somewhere i.e. in a database, on a server, in cloud storage, on a phone, in a backup, or on an encrypted hard drive.
Modern encryption usually depends on two broad types of cryptography:
Symmetric cryptography. Uses one secret key that locks and unlocks the data. It’s fast and works well for encrypting large amounts of information. AES is a common example. If a company encrypts a database, a backup file, or a hard drive, symmetric encryption may be involved.
Asymmetric cryptography. Often called public-key cryptography, uses two keys: a public key and a private key. The public key can be shared. The private key must be protected.
When your browser connects to a website, it needs a way to confirm that the website is really the website it claims to be. It also needs a way to set up an encrypted session without already sharing a secret key. Public-key cryptography helps make that possible.
Public-key cryptography is used in many places like TLS certificates on websites, VPNs, identity providers, API authentication, and some secure messaging.
The concern is that widely used public-key systems are built on math problems that normal computers struggle to solve, but a sufficiently powerful quantum computer could solve some of those problems much faster. In the wrong hands, that could allow an attacker to decrypt information that was supposed to stay private or impersonate systems that were supposed to be trusted.
Symmetric encryption like AES-256 is less exposed and generally considered more durable against known quantum attacks.
So the issue is not “encryption is dead.” The issue is that some of the trust systems that let companies, websites, devices, and software prove who they are may need to be replaced and that replacement will be messy because cryptography is everywhere.
It’s in browsers, servers, cloud platforms, VPNs, mobile apps, payment systems, and more. It’s in legacy applications that no one wants to touch because they still run an important business process and are prone to breaking the second anyone tries to modify them.
Most companies don’t have a clue where and what types of cryptography they use. That lack of visibility seems to be the biggest practical problem for preparing today.
Quantum Computing
Quantum computing is not artificial intelligence. That may seem obvious, but the two often get lumped together because they both sound futuristic and technical.
I recently finished a TV show where part of the plot involved someone creating a powerful quantum computer that became sentient and could see the future. I won’t go too far down the rabbit hole of how often we turn technology into a character with human motives, but it was a good example of how easily quantum computing and AI get blurred together in popular culture.
Quantum computing is a different model of computing based on quantum mechanics. A normal computer uses bits, which are represented as either 0 or 1. A quantum computer uses quantum bits, or qubits, which can behave in ways that allow certain problems to be approached differently.
Quantum mechanics is well above my pay grade, so I am not going to pretend I’m a physicist. The 101 is that quantum computers may eventually be able to solve certain specific problems much faster than today’s computers.
They will not be faster at everything. A quantum computer is not a supercomputer that will replace everything. Its advantage is much narrower than that.
One area where quantum computers may have an advantage is cryptanalysis, which is the study of breaking cryptographic systems.
In the 1990s, mathematician Peter Shor developed an algorithm showing that a sufficiently powerful quantum computer could solve certain math problems much faster than a normal computer. Those math problems are part of the foundation for widely used public-key encryption systems.
That is the issue here.
Today’s public-key encryption relies on math that is extremely difficult for normal computers to reverse. That difficulty is what makes the system useful. If a powerful enough quantum computer can solve those problems efficiently, then some of the encryption and authentication systems we rely on today would no longer be safe.
This doesn’t mean quantum computers are breaking encryption today, but it does mean the assumptions behind some widely used cryptography may eventually fail.
Encryption That Ages Badly
If a powerful quantum computer becomes available, attackers could potentially break systems that depend on specific types of cryptography.
As I stated, that isn’t possible yet, so the more immediate concern is called “harvest now, decrypt later.”
In that scenario, an adversary collects encrypted data today and stores it. The data may be unreadable now, but if the adversary believes they may be able to decrypt it in the future using quantum capabilities, the collection still has value.
This exposure depends heavily on the shelf life of the data, because some data has short lived value. A time sensitive notification that is only relevant before a specific date/time or a non-sensitive email may not matter a few years down the road. Other data retains value for a long time i.e. legal files, sensitive personal information, source code, trade secrets, financial records, and government communications.
For those categories, confidentiality could be important for decades. This is why the issue matters before quantum computers are fully mature. If data is being intercepted and stored now, the later arrival of quantum decryption capability could affect information that was compromised years earlier.
The second exposure is digital signatures. Encryption protects confidentiality, but digital signatures protect authenticity and integrity. They help prove that software, documents, transactions, certificates, and communications came from the expected source and were not altered.
If quantum computing undermines the public-key systems used for digital signatures, the issue becomes impersonating trusted entities, forging signatures, compromising software update mechanisms, or weakening certificate based trust.
That could affect software supply chains, financial systems, identity infrastructure, and entire device ecosystems.
There are ways to mitigate and migrate to more quantum resistant encryption, but that implementation takes time and with so much technology debt, it could break things and cause loads of friction.
A sober view starts with what we don’t know.
We don’t know exactly when a quantum computer capable of breaking modern public-key cryptography will exist. Estimates vary and many smart people think it’s still decades away while others think it could be before 2029. Either way, attackers can’t currently break modern cryptography at scale using quantum computers.
That matters because overstating the urgency can damage credibility. Companies already face active cyber threats like ransomware and business email compromise. Quantum risk shouldn’t move up the list of priorities ahead of any basic cyber hygiene. We still have lots of businesses that lack the basics like MFA on remote access and email, solid backups, and an IR plan, so post-quantum cryptography shouldn’t be the most urgent security project. One of the most important things you can do now is actually protect your encrypted data so it isn’t stolen then decrypted later.
It will take years for larger orgs to migrate to post-quantum cryptography. A good first step would be trying to itemize and classify your data. Answering questions like is it encrypted, what type of encryption, does this data have a long shelf value, etc.
From an insurance perspective, quantum risk is just now starting to be discussed and has accounted for zero losses. For certain classes of business, underwriters may eventually want to understand whether the applicant has begun addressing cryptographic risk. I imagine this will be most relevant for financial institutions, healthcare, defense contractors, technology companies, cloud providers, telecoms, MSPs, payment processors, law firms, and entities holding long shelf value confidential data.
Then again if the market conditions continue the way they have for the last few years, we’ll see lots of quantum related losses with no underwriting and pricing so low they’re basically giving it away.
I think the practical takeaway here is that quantum computing is not a reason to panic and you’ll likely see lots of the alarmist types start to create articles and webinars soon, but don’t automatically buy the doom narrative.
It’s hardly any different than other emerging exposures. The unprepared will get their teeth kicked in initially, then the market matures and quantum risk management becomes a priority and we start signaling the fire alarm about some new exposure.