ISO/IEC 42001 certifies an AI management system (AIMS): the policies, roles, risk and impact assessments, and lifecycle controls that govern how your organization builds, buys or uses AI.
It follows the same management-system structure as ISO 27001, so an existing ISMS gives you a large head start; the genuinely new work is AI risk, AI impact assessment and AI lifecycle control.
Customers, regulators and the EU AI Act are turning "responsible AI" from a statement into something you have to evidence. ISO 42001 is the most widely recognized way to do that today.
What ISO 42001 is
ISO/IEC 42001:2023, published in December 2023, is the first international standard that specifies requirements for an artificial intelligence management system, or AIMS. Like ISO 27001 for information security, it does not tell you which AI to build or how to build it. Instead, it defines the management system around your AI: who is accountable, how risks and impacts are assessed, which controls apply across the AI lifecycle, and how the whole system is monitored and improved over time.
The standard applies to any organization that develops, provides or uses AI systems, regardless of size or sector. That breadth matters. A SaaS company embedding a third-party large language model into its product is in scope just as much as a team training its own models. What changes between them is the scope of the AIMS and which controls are relevant, not whether the standard applies.
Certification is performed by an accredited certification body, following the familiar ISO pattern: a two-stage initial audit, a certificate valid for three years, and annual surveillance audits in between.
The structure: clauses 4–10 plus Annex A
ISO 42001 uses the harmonized structure shared by ISO 27001, ISO 9001 and other management-system standards. Clauses 4 to 10 are mandatory requirements: understanding your context and the AI systems in scope (4), leadership commitment and an AI policy (5), planning, including AI risk assessment, risk treatment and AI system impact assessment (6), support, covering resources, competence, awareness and documented information (7), operation (8), performance evaluation through monitoring, internal audit and management review (9), and continual improvement (10).
Annex A then provides 38 reference controls grouped under nine control objectives: policies related to AI; internal organization; resources for AI systems; assessing impacts of AI systems; the AI system life cycle; data for AI systems; information for interested parties; use of AI systems; and third-party and customer relationships. As with ISO 27001, you select the controls that address your risks and justify any you exclude in a Statement of Applicability.
Annex B gives implementation guidance for each control, and Annexes C and D describe common AI-related objectives, risk sources and how the standard applies across different domains. They are not requirements, but they are useful when you design your controls.
What is genuinely new compared with ISO 27001
If you already run an ISO 27001 ISMS, much of ISO 42001 will feel familiar. Governance, document control, competence, internal audit, management review and corrective action work the same way and can usually be shared across both systems. The real new work sits in three places.
First, AI risk assessment. Information security risk asks what could compromise the confidentiality, integrity or availability of information. AI risk is broader: it covers bias and unfair outcomes, lack of transparency or explainability, poor robustness, misuse, and over-reliance on automated decisions. You need a method that identifies and evaluates these risks consistently for each AI system.
Second, the AI system impact assessment. This is a distinct requirement to assess the potential consequences of an AI system for individuals, groups of individuals and society, not just for the organization. It forces the question many teams skip: who could be affected by this system, and how?
Third, AI lifecycle and data controls. Annex A expects defined processes for design, development, verification, validation, deployment, operation and monitoring of AI systems, and for the quality and provenance of the data they use. For organizations that consume third-party AI rather than building it, supplier and "use of AI systems" controls become the centre of gravity instead.
Who needs ISO 42001 now
Three groups feel the pressure first. AI product companies selling to enterprises are increasingly asked in security questionnaires and procurement reviews how they govern their models, and an ISO 42001 certificate answers that question in a form buyers recognize. Organizations using AI in sensitive decisions, such as hiring, credit, insurance, healthcare or education, need a defensible way to show they assessed and control the impacts on people. And companies with exposure to the EU AI Act need a management backbone for the obligations that apply to them.
On that last point, be precise. ISO 42001 certification does not by itself demonstrate compliance with the EU AI Act, and it is not currently a harmonized standard under the Act. But the Act expects risk management, data governance, documentation, human oversight and post-market monitoring, and an AIMS is the natural place to run all of them. Designing your AIMS with the Act in mind means one program produces evidence for both.
If none of these apply yet, ISO 42001 is still a sensible next step for any organization already certified to ISO 27001 that is putting AI into its products or operations. Adding it early is far cheaper than retrofitting governance once AI is embedded everywhere.
How long it takes and where to start
Timelines depend on how many AI systems are in scope and how mature your existing management system is. As an indication, organizations that already hold ISO 27001 often reach certification readiness for ISO 42001 in around three to six months, because the shared clauses carry over. Starting from scratch typically takes longer, as the core management system has to be built as well.
The practical first step is an inventory: list every AI system you develop, provide or use, including AI features inside third-party tools, and record its purpose, data and the people it affects. From there, a gap assessment against clauses 4 to 10 and the 38 Annex A controls shows what you can reuse and what you need to build.
If you already hold or are pursuing ISO 27001, assess both standards together. The management-system clauses are reviewed once, and the work focuses on what is specific to each: Annex A security controls for 27001, and AI risk, impact assessment and lifecycle controls for 42001. You end up with one roadmap and one integrated management system rather than two parallel programs.