Does quantum computing break all encryption?

No. Quantum computing threatens one family of encryption much more than the other, and knowing which is which turns an overwhelming problem into a specific one.

Most systems use two families together: public-key methods, which agree keys and create signatures, and symmetric methods, which encrypt the data itself with a shared key. A secure connection to a website typically uses both, one after the other.

Two families, two different risks

Public-key methods rely on mathematical problems that a capable quantum computer could solve efficiently. Symmetric methods don’t have that weakness; the best known quantum attack against them offers a much smaller advantage, which larger keys absorb.

That’s why quantum readiness is mostly about replacing public-key methods, not about re-encrypting everything you hold.

What changes

Public-key methods such as RSA and elliptic-curve cryptography are the concern. They’re used to agree keys between systems, to sign certificates and software, and to prove identity. NIST has proposed deprecating them after 2030 and disallowing them after 2035, in a draft that is still open.

The replacements are already standardized. NIST published its first three quantum-safe standards in August 2024: one for agreeing keys and two for signatures, described in plain words in the glossary.

What mostly doesn’t

NIST says its symmetric standards are “significantly less vulnerable to known quantum attacks” than its public-key standards. Symmetric methods with at least 128 bits of security, such as AES-128 and stronger, are believed to meet its lowest quantum-safe category.

There’s one caveat: the few symmetric options at a lower strength are due to be disallowed in 2030. Checking key sizes is part of the job, but it’s a small part compared with replacing public-key methods.

A worked example: one secure connection

Take an ordinary secure connection between a browser and a website. First, the two sides use a public-key method to agree a shared secret key, and the website proves its identity with a certificate, which is itself signed with a public-key method. Then they switch to a symmetric method, such as AES, to encrypt everything they send.

Against a capable quantum computer, the first part is the weak point: both the key agreement and the certificate signature use public-key methods. The symmetric encryption that protects the actual data holds up, provided its keys are long enough. Moving that connection to quantum-safe encryption means changing how keys are agreed and how identity is proven, while the data encryption stays much as it is.

Common misconceptions

• “All our encrypted data will be readable.” Data encrypted with adequate symmetric keys stays protected; the exposure is in how those keys were agreed and in signatures.
• “Longer RSA keys solve it.” NIST’s proposal covers RSA and elliptic-curve methods at every key size.
• “There’s nothing to do until the algorithms settle.” NIST’s first standards were published in 2024, and finding where public-key methods are used is needed whatever you choose.

What that means for where you start

The job is specific:

• Find where public-key methods are used, for agreeing keys, for signatures and for identity, and who owns each place.
• Plan their replacement in an order that doesn’t break what depends on them, starting with the data that must stay private longest and the signatures that must stay trusted longest.
• Check that symmetric key sizes meet the 128-bit minimum, and fix the few that don’t.

That’s still a multi-year programme for most organizations, but it’s a bounded one. Why plan for quantum-safe encryption now covers where to begin.