A password is not a key. This part makes the gap expensive.
The first edition of ISO/IEC 11770-8 is at final draft ballot. The public record names two functions, Argon2i and PBKDF2, and puts the rest behind a paywall. Check whether it reaches you in three questions, then read exactly what is public.
Free record and check. Kit USD 799 once. Watch USD 49 a month.
- Document
- ISO/IEC FDIS 11770-8
- Edition
- 1, a new part
- Committee
- ISO/IEC JTC 1/SC 27
- Stage
- 50.20, FDIS ballot, 8 weeks
- Stage date
- 21 September 2026
- Published
- Not yet. No date recorded.
days to the expected publication date, 16 February 2027. An estimate made for planning, not an ISO date and not a legal deadline.
What the document does, and what it does not do
It specifies key derivation functions built to take a human-memorised password as input. It applies where a cryptographic key has to come out of a password. That is narrower than the market assumes: a login hash is not in scope merely because a password went in.
The public introduction is candid about why the problem is hard. Passwords carry little entropy and can be guessed. Each guess can only be made expensive, never impossible. The two named functions, Argon2i and PBKDF2, are the document's answer to that economics.
What this site is, and is not
It is a reading of the public record: the ISO catalogue page, the OBP public preview, and the public specifications of the functions the draft names. It is not the standard, not a reproduction of it and not a compliance result.
The normative clauses and the four annexes are behind the paywall. Every page says so. Not legal advice; does not certify.
Three questions. One reading. No account.
What the reading uses
Your answers are matched against clause 1 (scope), the two functions named in the public introduction, and the evidence a reviewer will ask for. The date is the estimate, labelled.
It never calls an answer compliant. The five phrases are: in scope with gaps, in scope and holding, likely in scope, out of scope on your answers, and scope incomplete.
Need the salt, inventory and ISMS questions too? Run the full check. It adds four questions and a longer plan.
What is not public, and why we will not invent it
Only the informative sections are public: the foreword, the introduction, clauses 1 to 3, the definitions and the contents. The rest is locked. That is the central fact a buyer needs, and most pages about a draft omit it.
- Clauses 4 to 7: the symbols, the use clause, Argon2i and PBKDF2.
- Annexes A to D: object identifiers, the compression function G, Blake2b, and parameter guidance.
- The figures, the identifiers and every normative "shall".
- Any publication date. ISO records none.
So the register on this site does two things. Where the public record proves a point, it states it and cites the clause. Where it does not, it says "not stated in the public preview" and gives our reading, labelled as a reading. A page that cannot tell those apart is not worth your time.
The one correction worth making first
Annex A is not a control set. It assigns object identifiers for the algorithms. There is no Annex A control list to audit against, and this part sets no management system requirements. If a supplier tells you otherwise, ask for the clause.
The stage history, with the blanks left blank
| Stage | What it means | Date | Record |
|---|---|---|---|
| 10.99 | New project approved | 2024-01-05 | confirmed |
| 20.00 | New project registered in TC/SC work programme | 2024-01-05 | confirmed |
| 30.00 | Committee draft registered | 2024-12-04 | confirmed |
| 30.20 | CD consultation initiated | 2024-12-05 | confirmed |
| 30.60 | Close of comment period | 2025-01-31 | confirmed |
| 30.99 | CD approved for registration as DIS | 2025-06-05 | confirmed |
| 40.00 | DIS registered | 2025-07-21 | confirmed |
| 40.20 | DIS ballot initiated: 12 weeks | 2025-09-22 | confirmed |
| 40.60 | Close of voting | 2025-12-16 | confirmed |
| 40.99 | Full report circulated: DIS approved for registration as FDIS | 2026-06-24 | confirmed |
| 50.00 | Final text received or FDIS registered for formal approval | 2026-06-25 | confirmed |
| 50.20 | Proof sent to secretariat or FDIS ballot initiated: 8 weeks | 2026-09-21 | confirmed |
| 50.60 | Close of voting, proof returned by secretariat | not recorded | no date |
| 60.00 | International standard under publication | not recorded | no date |
| 60.60 | International standard published | not recorded | no date |
Source: ISO catalogue page 87940, retrieved 2026-10-21. The 50.20 date is 21 September 2026; the ballot runs eight weeks, which we calculate as a close on 16 November 2026. The full date page and plan.
Both are named. Neither is a default.
Argon2i
Argon2i makes each guess cost memory as well as time. Its addressing does not depend on the password, which is why it is chosen on shared or virtualised hosts where a cache-timing observer is plausible. The trade is a weaker defence against some time-memory trade-offs than the data-dependent variant.
- Named in
- Introduction; clause 6
- Parameters
- Memory, iterations, parallelism, salt, output length
- Evidence
- Parameter record, benchmark, test vector for this variant
PBKDF2
PBKDF2 raises the cost through an iteration count and a pseudorandom function. It is available in validated modules and in the libraries most teams already run, which matters in a regulated estate. The count ages with hardware, so the benchmark needs a date and a review trigger.
- Named in
- Introduction; clause 7
- Parameters
- Iterations, salt, PRF, output length
- Evidence
- Parameter record, benchmark date, test vector with the production PRF
Argon2i and PBKDF2 are public specifications as well as ISO clauses. This site explains them from their public specifications and never presents that explanation as the ISO text. The longer comparison.
Eleven units, each with its public status
Pay for the work, not for the standard text
The register, the evidence list and the dated plan as PDF, CSV and XLSX, plus the parameter record template and the questionnaire answer pack.
- 11 units, 38 evidence items, every field sourced
- Dated plan with the estimate separated from any deadline
- Parameter record and decision record templates
- Questionnaire answer pack for supplier reviews
An email when the record changes: a stage date, a new contents entry, a corrigendum, a related part, or the published text.
- The diff, in plain words, with the source link
- What it changes in the register and the date
- Cancel any time from the Stripe receipt
- No newsletter, no marketing list
One click to Stripe. No account gate. The kit is served only against a paid Stripe session. The price is in the page, in the schema and in llms.txt, in the same words. What is included, in full.
The questions buyers ask first
- Is ISO/IEC 11770-8 published yet?
- No. On 21 October 2026 ISO records it at stage 50.20, the final draft ballot, with 22 pages. The expected publication date of 16 February 2027 is our estimate, not an ISO date.
- Does it replace another part of the series?
- No. It is the first edition of a new part. It adds password-based key derivation to a series that already covers frameworks, symmetric and asymmetric mechanisms, weak secrets, group key management and cross-domain password authentication.
- Which functions does it specify?
- The public introduction names Argon2i and PBKDF2. It does not name Argon2id, scrypt or bcrypt, so a team using another function has a decision to record.
- Does ISO/IEC 27001 require it?
- Not by name. The 2022 cryptography control requires a policy and a key management approach; it does not cite this part. A contract, a regulator or a scheme can make the part binding.
- Is login password hashing in scope?
- Not on the face of clause 1. The scope turns on deriving a key for another algorithm or application. A hash used only to verify a password is a different job.
- What does the kit contain?
- The 11-unit obligation register with 38 evidence items, the dated plan, a parameter record template, a questionnaire answer pack and the reading guide, as PDF, CSV and XLSX.