ISO/IEC 42001 AIMS Scope & Context Builder
Your AIMS scope says which activities, AI systems, business units and sites your AI management system governs, and why anything is left out. Work through six steps (organisation, AI roles, issues, interested parties, AI systems, boundaries) and get a draft scope statement, a context register and a list of inclusions and exclusions, with the questions a certification auditor would ask already flagged.
Loaded with a sample for a fictional health-tech company. Saved in your browser only. A template, not certification advice.
You are looking at a sample for a fictional health-tech company, with a few deliberate gaps for the checks to catch. Edit anything, or . Your work is saved in this browser only.
1. Organisation profile
No auditor flags on this step.
TL;DRWhat this builder does
5 points
- ISO/IEC 42001 clause 4 asks you to understand your context (4.1), the needs of interested parties (4.2), and then fix the scope of the AI management system (4.3).
- Six guided steps: organisation profile, AI roles, external and internal issues, interested parties and their requirements, AI systems in or out of scope, and boundaries and interfaces.
- The output is a draft scope statement, a context register of issues and interested parties, and a list of inclusions and exclusions with reasons.
- Auditor-style checks flag what a certification auditor would question: an exclusion with no justification, a jurisdiction with no regulator among interested parties, an outsourced process left out of scope.
- Free to use and saved in your browser. Markdown and CSV export are part of Pro.
AIMS scope vs an AI security scope
This is a management-system scope. It defines the organisational boundary of the ISO/IEC 42001 AIMS: who is governed, which AI systems, which sites and units, and which outsourced processes. It is the scope a certification body audits and prints on the certificate.
The ISO/IEC 27090 artefacts include a different kind of scope statement: the security scope for AI systems under ISO/IEC 27090, which is guidance and cannot be certified. It lists the models, data, pipelines and threats your AI security work covers. Write the AIMS scope first, then make sure the 27090 security scope sits inside it.
What auditors test in clause 4
At Stage 1 the auditor reads the scope and context; at Stage 2 they test it against reality. Common findings: exclusions justified only as "not relevant", AI systems customers use left out of scope, an interested-party list with names but no requirements, regulators missing for markets the company sells into, and outsourced model APIs treated as somebody else's problem. The checks in this builder look for exactly those patterns.
Where to go next
Feed the in-scope systems from your AI system inventory, run an ISO/IEC 42005 impact assessment for systems that affect people, then decide which Annex A controls apply in the Statement of Applicability builder.
Frequently asked questions
What does ISO/IEC 42001 clause 4 require?
Clause 4, context of the organisation, asks you to determine the external and internal issues relevant to your purpose and to the intended outcomes of the AI management system, including your role with respect to AI systems (4.1); the interested parties, their relevant requirements and which of those requirements the AIMS will address (4.2); and the boundaries and applicability of the AIMS, written down as its scope (4.3). Clause 4.4 then requires you to establish, implement, maintain and improve the AIMS itself.
What should an AIMS scope statement contain?
It should name the organisation, the AI roles it plays (provider, developer, deployer or user), the AI systems and activities covered, the business units and sites involved, outsourced processes that sit inside the AIMS, the interfaces with other management systems, and every exclusion with its justification. Certification bodies print a shortened version on the certificate, so it has to be precise.
How is this different from the ISO/IEC 27090 scope statement?
The AIMS scope is a management-system scope: it says which parts of the organisation and which AI systems the ISO/IEC 42001 management system governs, and it goes on the certificate. The scope statement in the ISO/IEC 27090 artefacts is a security scope for AI systems under guidance that cannot be certified: which models, data and pipelines the AI security work covers. They should agree with each other, but they answer different questions.
Can I exclude AI systems or business units from the AIMS scope?
Yes, but each exclusion needs a justification and must not undermine the AIMS outcomes. Auditors challenge exclusions of AI systems you provide to customers, units that can still deploy AI into production, and outsourced processes. Outsourcing a process does not remove it from the AIMS; externally provided processes still have to be controlled.
Why does the tool flag a jurisdiction with no regulator?
If you operate in, or affect people in, a jurisdiction, its regulators are interested parties with requirements, such as data protection authorities or AI-specific authorities. A context register that lists the EU but no EU authority suggests the legal requirements were not analysed. The hints name well-known bodies only as prompts; you decide which regulator actually applies.
Where is my data stored, and is this free?
Everything is saved in your own browser with localStorage; nothing is sent to PlayCISO, and clearing site data or using a private window removes it. The guided steps, draft statement, context register and checks are free. Downloading the pack as Markdown or CSV is part of Pro ($5/month or $40/year). This is a template, not certification advice, and it does not reproduce the standard.
More security tools
The tools people use most on PlayCISO.
Catch injected instructions hiding in content your AI agent should only read.
Score your readiness against the first international standard for AI security threats.
Critical CVEs across 21 major vendors, month by month — and what the 2026 surge means.
Build an ordered ruleset, replay labelled packets and catch shadowed and any-to-any rules.
Self-assess all six Functions, from Govern to Recover, and see your weakest one.
Search 172 threat actors across Microsoft, CrowdStrike, Mandiant and community names.