Control the allocation and use of secret authentication information, ensuring appropriate secrecy is maintained.
Also written as A5.17, Annex A 5.17, ISO 27001:2022 A.5.17, ISO27001 A.5.17.
Authentication information is one of 37 Organisational Controls in ISO/IEC 27001:2022 Annex A. Control the allocation and use of secret authentication information, ensuring appropriate secrecy is maintained. Organisational controls are judged on governance rather than tooling: an auditor wants a named owner, an approval trail, and evidence the control is exercised on a defined cadence rather than written once and filed.
When preparing your Statement of Applicability (SoA) for A.5.17, gather artefacts such as:
If you run more than one framework, the same evidence usually satisfies all of them. A.5.17 aligns with:
SOC 2: CC6.1 Logical access security architecture, CC6.2 Provisioning and deprovisioning
NIST CSF 2.0: PR.AA-01 Identity and credential management, PR.AA-03 Authentication of users, services and hardware
A.5.17 consolidates the following ISO 27001:2013 control(s): A.9.2.4, A.9.3.1, A.9.4.3. If you are transitioning an existing ISMS, map your prior evidence for these to A.5.17 in your updated SoA.
See all 37 Organisational Controls →
Annex A controls are not mandatory in the abstract. Clause 6.1.3 requires you to compare your risk treatment plan against Annex A and justify, in the Statement of Applicability, any control you exclude. If your risk assessment surfaces a risk that A.5.17 addresses, excluding it needs a documented, risk-based rationale that an auditor will test.
In two passes. First design: does a documented control exist, is it owned, and does it address the risk? Then operating effectiveness: the auditor samples records from across the audit period to confirm the control actually ran. A Stage 2 audit will typically pull several samples, so evidence that only exists for the month before the audit is a common finding.
ISO 27001 sets no fixed interval — it requires review at "planned intervals" and after significant change. Annual review is the norm most certification bodies expect, with an out-of-cycle review triggered by incidents, major system changes, restructures, or new regulatory obligations. Record the review date and outcome either way; an undated control is treated as unreviewed.
A.5.17 aligns with SOC 2 CC6.1, CC6.2 and NIST CSF PR.AA-01, PR.AA-03. Evidence gathered for one framework will usually satisfy the others, which is the basis for a test-once, satisfy-many control library.
A.5.17 consolidates 3 control(s) from the 2013 edition: A.9.2.4, A.9.3.1, A.9.4.3. When transitioning, re-point the existing evidence rather than rebuilding it — the underlying requirement has not changed materially.