Suneet Singal on post-quantum planning in plain language

Suneet Singal’s article library includes a preserved essay on post-quantum security, and this page is the plain-language companion rather than a second copy of it. Some of the public-key methods used to set up encrypted connections and to sign documents were designed for today’s computers. A sufficiently capable quantum computer would change the outlook for certain of those methods. In August 2024 the National Institute of Standards and Technology finalized three standards so organizations can plan against that possibility with named tools. Planning here means an inventory and a question about how long information must stay secret or trustworthy. It does not mean a timetable this page invented, and it does not mean instructions for an attack. Suneet Singal’s published background is real estate and corporate finance. The entry is a reading guide for a topic already in the library, not a claim that he operates a migration and not personal security advice.

The planning problem in ordinary words

A great deal of everyday security depends on two jobs. One job is agreeing on a secret that will protect a conversation or a stored file, even though the agreement itself happens over a channel other people can see. The other job is signing something so that a recipient can check who signed it and whether it was altered. Those jobs are often done with public-key methods. The methods are public in the sense that one half of the key can be shared, and they are hard to break with today’s computers when they are used correctly. The planning problem is that a future machine, of a kind researchers describe as a cryptographically relevant quantum computer, would not be stopped by some of those methods in the same way.

The defensive reason to think about this before such a machine exists is the lifetime of the information. Encrypted material can be recorded now and kept. If the method protecting it later falls, the recording becomes interesting to whoever kept it. People call that pattern harvest-now, decrypt-later. This page uses the name as a planning phrase. It does not describe how anyone would record traffic, which systems they would choose, or how a future decryption would be attempted. The only operational content is the question a defender asks: how long must this information remain confidential, and which public-key method is protecting it today?

Signatures raise a related but different question. A signature that must still be trustworthy years from now, on software, on a recorded document, or on a long-lived certificate, is a planning object even when confidentiality is not the issue. Some signatures only need to be trustworthy for a short session. The inventory below is how you tell those cases apart. Suneet Singal’s preserved article, “Post-Quantum Security Is No Longer a Future Problem,” at /post-quantum-security-is-no-longer-a-future-problem/ , discusses the topic at its own date and in its own words. Read that date on the article page. This entry does not copy the article, does not silently update it, and does not adopt every timeline a reader might remember from it. Standards and guidance move. The editorial standards page explains why a preserved date still governs the preserved text.

What NIST finalized in 2024

In August 2024 NIST announced the first three finalized post-quantum standards. FIPS 203 defines ML-KEM, a module-lattice key-encapsulation mechanism, for the job of establishing a shared secret. FIPS 204 defines ML-DSA, a module-lattice digital signature algorithm. FIPS 205 defines SLH-DSA, a stateless hash-based digital signature algorithm. The announcement and the project page are the public sources. The finalized publication pages for FIPS 203, FIPS 204, and FIPS 205 are the documents themselves. A reader who needs the parameter sets, the security claims, or the implementation requirements should read those documents. This entry stops at the names and the jobs, because the names are what a planning conversation needs and the parameter sets are what an implementer needs.

Key encapsulation and digital signature are the two jobs to keep straight. ML-KEM is for establishing keys. ML-DSA and SLH-DSA are for signatures, and they are two different signature families, not two brand names for one product. A sentence that says “use the NIST algorithm” has not yet said which of the three standards, or which job. A sentence that says “quantum encryption” has usually mixed this work up with quantum key distribution, which is a different approach that uses quantum hardware to distribute keys. The standards NIST finalized in 2024 are classical algorithms, meant to run on ordinary computers, designed so that a future quantum computer does not break them in the way it would break the older public-key methods they are intended to replace in those roles.

NIST’s project page is also where a reader checks what else is still in process. This entry will not freeze a list of draft algorithms or a promise about a fourth standard, because drafts move. If a product sheet cites a draft, the honest note says “draft, as of the date on the sheet,” and the project page is how you see whether that draft is still the one NIST is discussing. Finalized, in the August 2024 announcement, applies to the three standards named above. It does not apply to every slide that uses the words post-quantum.

What the three names do not cover

The three standards do not mean that every cipher in a building has been replaced, or that every cipher needs the same replacement. They address public-key key establishment and signatures. Symmetric encryption, the kind that uses a shared secret key to protect the contents of a message after the secret has been agreed, is a different family. A planning note that marks every use of encryption as “broken by quantum” has flattened a distinction NIST’s own materials exist to preserve. Read the project page before you write that note. The defensive habit is to record which algorithm is in use and what job it is doing, not to apply one adjective to the whole stack.

The standards also do not install themselves. A FIPS publication is a specification. Vendors, libraries, certificate authorities, devices, and applications adopt specifications on their own schedules, and some products will speak a hybrid of an older method and a newer one during a transition. This page does not rank vendors and does not declare a hybrid mandatory or forbidden. It says that a purchase or a design should name the standard, the job, and whether the deployment is hybrid, and that the name should be something you can find on the NIST page. A brochure that says “quantum-safe” without a FIPS number has not yet given you a row for the inventory.

Nothing here is a deadline. Agencies and standards bodies publish migration materials, and those materials can contain dates. The Cybersecurity and Infrastructure Security Agency maintains a post-quantum cryptography initiative, and NIST maintains the project page. If a date matters, read it on those pages on the day you are planning, and copy the date with the document title. Do not take a year from memory, from the preserved article, or from this entry. This entry refuses to add one. A wrong year in a reference is worse than a pointer, because it looks like the kind of fact the rest of the page was careful about.

Data lifetime is the sorting question

Data lifetime is the period during which confidentiality or authenticity still matters. A session that protects a catalog page for a few seconds does not present the same planning question as a file that must stay confidential for many years, or a signature that must still verify on a document a decade from now. The inventory asks the owner to say which of those is true, in words the owner will stand behind. It does not ask the owner to predict the day a quantum computer arrives. CISA’s and NIST’s pages discuss the planning context. The owner’s page discusses the information. The comparison between those two is the planning act. This essay will not replace that comparison with a slogan about acting now or waiting.

Lifetime also differs inside one system. A messaging application may protect a live conversation and also store backups. The conversation and the backup can have different lifetimes and different algorithms. A database column with health information and a database column with a public store address should not share a row merely because they share a server. The point of an inventory is to split rows until the lifetime and the algorithm are honest. A single row called “the network” is a wish that the problem were smaller.

If you do not know the lifetime, write “unknown” and assign someone to ask. Unknown is a finding. It means you cannot yet decide whether the harvest-now concern is even relevant to that row. Guessing a long lifetime for everything produces a plan that cannot be funded. Guessing a short lifetime for everything produces a plan that ignores the files that actually need it. The cryptographic-inventory entry on this site goes further into the columns. This page contributes the reason the lifetime column exists.

An inventory comes before a shopping list

The defensive sequence is to list where public-key cryptography is used, to attach a lifetime and an owner, and only then to talk about which NIST standard fits which job. Shopping first produces a pile of product claims. The proptech entry is about product claims in real estate. The same reading habit applies here. A claim that a firewall, a laptop, or a cloud service is “post-quantum” should be asked: which FIPS standard, which job, key establishment or signature, and what still uses an older method beside it? If the vendor cannot answer, the claim is an adjective.

The list covers systems the organization operates and services it buys. Certificates on websites, virtual private networks, device management, code signing, document signing, identity providers, and backups are ordinary places public-key methods appear. This page does not tell anyone to scan a network, to probe a system they do not operate, or to extract keys. The source of a row is documentation the owner is authorized to read: a vendor’s security paper, a configuration screen the owner administers, a certificate the owner can see in a browser they use for their own site, or a standard named by NIST. An inventory built by poking at other people’s systems is not the inventory this reference is describing.

Suneet Singal is not presented as having completed such an inventory for a named company, and the biography’s real estate and corporate finance record is not a security certification. The preserved article is journalism in the library, with its own date. This explainer is a vocabulary sheet so that article, the NIST announcement, and a vendor brochure can be read against the same words. If the words and a current CISA publication diverge, follow CISA and NIST. Then correct the vocabulary, which is allowed to improve, rather than treating this page as a procedure.

What this page will not schedule

It will not schedule a migration, name a year by which older methods must be gone, or tell a reader which of their systems to replace first. Priorities depend on the inventory and on guidance from NIST and CISA that the reader should fetch rather than remember. It will not describe attacks, algorithms for breaking keys, or the internal parameters of a quantum computer. It will not say that a particular bank, hospital, or house has already been compromised, or that a reader’s password is the thing post-quantum standards replace. Passwords and passkeys are an identity topic. The passkeys entry keeps them separate on purpose. They can appear in the same inventory as a row about authentication. They are not FIPS 203.

It will not claim that final standards ended the work. Final standards start a long period of implementation, testing, and replacement. Crypto agility, the ability to change algorithms without rebuilding an entire product, is a design idea a reader may meet in NIST’s discussion. This page names it so the phrase is not mysterious. It does not specify an architecture. Engineers who implement ML-KEM or the signature standards need the FIPS documents and current implementation guidance, not an essay on a personal site.

Readers who want a deeper treatment can read the preserved article and then check every time-sensitive sentence against the NIST and CISA pages, using the article’s publication date as the context. That is the archive-date habit applied to security writing. A sentence can have been reasonable on its publication date and still be the wrong sentence to paste into a plan drafted later. The plan should cite the standard, not the mood of an older headline.

A short vocabulary to keep

Public-key cryptography does the agreement and the signature jobs that started this page. A cryptographically relevant quantum computer is the future machine that changes the outlook for certain of those methods. Harvest-now, decrypt-later is the reason long-lived confidentiality cannot wait on a public demonstration. FIPS 203 ML-KEM establishes keys. FIPS 204 ML-DSA signs. FIPS 205 SLH-DSA signs with a hash-based method. An inventory lists where the older methods sit. Data lifetime says how long the row matters. CISA and NIST publish the current initiative and the project. The preserved article is related reading with its own date. None of these sentences is a configuration, a purchase order, or personal security advice.

If you write the vocabulary down, leave a blank line under it for the retrieval date of the NIST announcement and the CISA page. The announcement was August 2024. The initiative page is current only on the day you read it. That difference is the whole reason a planning vocabulary and a preserved article are not the same document. Use both. Do not let either one pretend to be a work order.

What this entry does not claim

Suneet Singal’s published professional background is in real estate and corporate finance. Company filings linked from this site record a chief executive officer departure from FC Global Realty Incorporated effective 2 January 2018 and an acting chief executive officer appointment at First Capital Investment Corporation effective 12 September 2017. The Suneet Singal Scholarship, Grant and Giving websites describe him as the founder of those programs. This entry explains a general subject. It is not a private track record, a current job title, an investment result, or a quote.

Continue with a cryptographic inventory, passkeys and passwords, and why an article’s date is part of its meaning.