Birbal AI

Guide · AI governance

ISO/IEC 42001 in the UAE: AI governance, certification and the rules that apply

ISO/IEC 42001 is the international standard for an AI management system, and the one an organization can be certified against. This guide explains what it requires, how certification works, how it lines up with the UAE's AI charter, the data protection law and sector rules, and how to prepare. The facts come from ISO's own pages and UAE official sources, checked on 18 November 2026.

Updated · 10 min read · By the Birbal AI team, Abu Dhabi · اقرأ بالعربية

Key takeaways

  • ISO/IEC 42001:2023 sets the requirements for an AI management system: policies, roles, risk and impact assessments, and controls, for any organization that develops, provides or uses AI. It was published in December 2023, and accredited certification bodies certify against it.
  • Its clauses follow the same structure as ISO/IEC 27001, and Annex A lists 38 controls in nine areas, from AI policy and impact assessment to the data used by AI systems and supplier relationships.
  • When we checked on 18 November 2026, we found no UAE law or regulation that requires ISO/IEC 42001 certification. The standard does give structure and evidence for what UAE rules ask: the UAE AI Charter's principles, impact assessments under the data protection law, the Central Bank's AI guidance and the DIFC's rules for autonomous systems.
  • In June 2026 the DIFC proposed a new Regulation 11 that would let its Commissioner of Data Protection recognise certification frameworks, and the UAE announced an Artificial Intelligence and Data Authority.

What ISO/IEC 42001 is

ISO/IEC 42001:2023, published by ISO and IEC in December 2023, is the first international standard for an artificial intelligence management system (AIMS): the policies, objectives and processes an organization uses to develop, provide or use AI responsibly. It applies to any organization, whatever its size or sector, that provides or uses products or services built on AI systems.

It is a management system standard, like ISO/IEC 27001 for information security. It says what the organization must have in place rather than how a particular model must work, and an organization can be audited and certified against it. Three related standards help: ISO/IEC 23894 gives guidance on AI risk management, ISO/IEC 42005:2025 gives guidance on AI system impact assessments, and ISO/IEC 42006:2025 sets the extra requirements for bodies that audit and certify AI management systems.

What the standard requires

ClauseWhat you doWhat an auditor looks for
4 ContextDecide the scope of the AI management system and your role for each AI system: developer, provider or user. Identify interested parties and what they require.A written scope, the AI systems in it and your role for each.
5 LeadershipTop management sets an AI policy, assigns roles and responsibilities, and gives the system the resources it needs.An approved AI policy and named owners.
6 PlanningAssess AI risks and decide how to treat them, assess the impact of AI systems on individuals, groups and society, and set AI objectives.Risk and impact assessments, a risk treatment plan and a Statement of Applicability against Annex A.
7 SupportProvide resources, competent people, awareness, communication and controlled documents.Training records and document control.
8 OperationCarry out the risk treatment, and repeat the risk and impact assessments at planned intervals and when a system changes significantly.Records of assessments and of the controls in use.
9 Performance evaluationMonitor and measure, run internal audits and hold management reviews.Audit reports and review minutes.
10 ImprovementCorrect nonconformities and keep improving the system.Corrective actions, with evidence that they worked.

Annex A lists 38 reference controls in nine areas. You decide which apply, and record the reasons for any you leave out:

  • A.2 Policies related to AI.
  • A.3 Internal organization: roles and responsibilities, and reporting of concerns.
  • A.4 Resources for AI systems: data, tools, systems and computing, and people.
  • A.5 Assessing impacts of AI systems on individuals, groups and society.
  • A.6 AI system life cycle: from objectives and requirements to design, verification, deployment, operation, monitoring, technical documentation and event logs.
  • A.7 Data for AI systems: acquisition, quality, provenance and preparation.
  • A.8 Information for interested parties: documentation for users, external reporting and communication of incidents.
  • A.9 Use of AI systems: processes and objectives for responsible use, and intended use.
  • A.10 Third-party and customer relationships: suppliers, customers and how responsibilities are shared.

Annex B gives implementation guidance for each control, Annex C lists possible AI objectives and risk sources, such as fairness, security, transparency and the level of automation, and Annex D explains how the system applies across domains and sectors.

How it fits the UAE's rules on AI

Rather than through one AI act, the UAE governs AI through a national charter, the data protection laws and sector rules. ISO/IEC 42001 does not replace any of them, but its policies, assessments and records are the evidence they ask for:

RuleWho it coversWhere ISO/IEC 42001 helps
UAE Charter for the Development and Use of Artificial Intelligence (July 2024)AI developed or used in the UAETwelve principles, among them safety, algorithmic bias, data privacy, transparency, human oversight, and governance and accountability. Annex A controls turn them into owners, assessments and records.
Federal Decree-Law No. 45 of 2021 on personal data (PDPL)Controllers and processors of personal data, except where Article 2(2) leaves them out: government data, health and banking data with their own laws, and free zones with their own data protection lawAn impact assessment before high-risk processing with new technology (Article 21), a data protection officer for such processing (Article 10) and the right to object to automated decisions (Article 18). AI impact assessments under A.5 can cover the same ground.
Central Bank guidance note on AI and machine learning (February 2026)Banks, insurers and other licensed financial institutionsThe Central Bank says it reinforces governance and transparency, human oversight and data protection in the use of AI. A working AI management system shows how these run day to day.
DIFC Data Protection Regulation 10 (2023)DIFC entities that deploy or operate autonomous and semi-autonomous systems processing personal dataClear notices, accountability and limits on high-risk processing by such systems. In June 2026 the DIFC proposed a Regulation 11 that would let its Commissioner recognise accreditation and certification frameworks; the consultation closed on 18 July 2026.
Federal law on AI in health (being drafted)The health sectorIn May 2026 the Cabinet approved drafting a federal law on smart health applications and AI in health, covering health data governance, safety and quality standards, liability and patient rights.

In June 2026 the UAE also announced the Artificial Intelligence and Data Authority, which brings together the AI Office, the UAE Data Office and the TDRA's Digital Government Sector. Federal entities are moving services to agentic AI (see our guide to agentic AI in the UAE), which makes an inventory of AI systems and clear accountability more pressing.

When we checked on 18 November 2026, we found no UAE law or regulation that requires ISO/IEC 42001 certification. Check your regulator's current rules and your contracts: buyers, especially government entities and banks, may ask suppliers for evidence of AI governance.

How certification works

A certification body certifies against ISO/IEC 42001; ISO itself does not. ISO/IEC 42006:2025 sets the extra requirements for bodies that audit and certify AI management systems, building on ISO/IEC 17021-1, the general standard for management system certification bodies.

  • Choose a certification body whose accreditation covers ISO/IEC 42001, and check it on the accreditation body's public register.
  • Stage 1 audit: the auditor reviews your scope, documents and readiness.
  • Stage 2 audit: the auditor checks that the system works in practice, from risk and impact assessments to the controls in your Statement of Applicability.
  • As with other ISO management system certificates, expect a certificate that runs for three years, with surveillance audits in between and a recertification audit before it ends. Confirm the cycle with your certification body.

If you already hold ISO/IEC 27001, much of the system carries over. Both standards share the same clause structure, so the policy framework, document control, internal audit and management review can be extended rather than rebuilt.

Birbal AI is not a certification body and does not issue ISO/IEC 42001 certificates.

How to prepare: seven steps

  1. List your AI systems

    Record every AI system you develop, provide or use, including AI features inside software you buy and AI agents, with an owner, its purpose, the data it uses and your role for each.

  2. Set the scope and the policy

    Decide what the AI management system covers, write the AI policy and name who is accountable, including an Autonomous Systems Officer in the DIFC where Regulation 10 calls for one.

  3. Assess risks and impacts

    Run AI risk assessments and impact assessments on individuals, groups and society, using ISO/IEC 23894 and ISO/IEC 42005 as guides, and combine them with PDPL impact assessments where personal data is involved.

  4. Choose your controls

    Go through the 38 Annex A controls, decide which apply and record the reasons in a Statement of Applicability.

  5. Put the controls to work

    Quality and provenance checks for training and test data, human oversight of decisions with real consequences, logs of what each system did, and contract terms with model and AI service suppliers.

  6. Check and improve

    Monitor the systems, run an internal audit and a management review, and fix what they find.

  7. Bring in the certification body

    Once the system has run long enough to produce records, book the Stage 1 and Stage 2 audits.

Where Birbal AI fits

Birbal AI builds and runs AI and data systems for organizations in the UAE. Warsha's AI agent governance keeps the records an AI management system needs: agent definitions with owners and versions, test results, approvals and a trace of every run, on your own servers. Warsha does not make you compliant on its own. Our data governance and PDPL compliance services cover the data behind your AI systems and the impact assessments the data protection law requires. Birbal AI is not a certification body; certification comes from an accredited certification body you choose.

Questions and answers

What is ISO/IEC 42001?

The international standard for an artificial intelligence management system, published by ISO and IEC in December 2023. It sets requirements for the policies, roles, risk and impact assessments and controls an organization uses to develop, provide or use AI responsibly.

Is ISO 42001 mandatory in the UAE?

When we checked on 18 November 2026, we found no UAE law or regulation that requires it. UAE rules do ask for what it provides, such as impact assessments, human oversight and accountability, and buyers may ask suppliers for evidence of AI governance.

How many controls are in ISO 42001?

Annex A lists 38 controls in nine areas, numbered A.2 to A.10. You decide which apply and record the reasons in a Statement of Applicability.

Who can certify an organization to ISO 42001?

A certification body accredited for ISO/IEC 42001. ISO/IEC 42006:2025 sets the extra requirements for bodies that audit and certify AI management systems. ISO itself does not certify.

How is ISO 42001 different from ISO 27001?

ISO/IEC 27001 manages information security risks. ISO/IEC 42001 manages the risks and impacts of AI, such as bias, transparency, human oversight and the data used to train and test models. Both use the same clause structure, so they can run as one integrated management system.

Does ISO 42001 cover generative AI and AI agents?

Yes. It applies to any AI system an organization develops, provides or uses, whatever the technique, including generative AI and agents that take actions.

What is an AI system impact assessment?

An assessment of how an AI system and its foreseeable uses may affect individuals, groups and society. ISO/IEC 42001 requires one, and ISO/IEC 42005:2025 gives guidance on how to do it.

Does the UAE Central Bank regulate AI?

In February 2026 the Central Bank of the UAE issued a guidance note on consumer protection and the responsible adoption and use of AI and machine learning by licensed financial institutions, including insurers. It emphasises governance, transparency, human oversight and data protection.

What does the UAE AI Charter require?

The UAE Charter for the Development and Use of Artificial Intelligence (July 2024) sets twelve principles, among them safety, addressing algorithmic bias, data privacy, transparency, human oversight, governance and accountability, and compliance with treaties and local laws.

How long is an ISO 42001 certificate valid?

As with other ISO management system certificates, a certificate normally runs for three years, with surveillance audits in between and a recertification audit before it expires. Confirm the cycle with your certification body.

Sources

  1. ISO/IEC 42001:2023, AI management system (ISO)
  2. ISO/IEC 42005:2025, AI system impact assessment (ISO)
  3. ISO/IEC 42006:2025, requirements for bodies certifying AI management systems (ISO)
  4. The UAE Charter for the Development and Use of Artificial Intelligence (ai.gov.ae, July 2024)
  5. Federal Decree-Law No. 45 of 2021 on the Protection of Personal Data (UAE Legislation)
  6. CBUAE issues guidance note on the responsible use of AI in the financial sector (Central Bank of the UAE)
  7. DIFC enacts amended data protection regulations, including Regulation 10 (WAM, September 2023)
  8. DIFC announces consultation on amended DIFC Data Protection Regulations (DIFC release, 18 June 2026)
  9. UAE Cabinet meeting of 18 May 2026: agentic AI and AI in health (Dubai Media Office)
  10. Mohammed bin Rashid approves establishing Artificial Intelligence and Data Authority (WAM, 14 June 2026)

Birbal AI is an independent company based in Abu Dhabi. Facts about third parties come from the sources above; check them for the latest position.

Book a demo

See the records an AI management system needs, on a live agent.

We build a small agent, restrict its tools, run an evaluation, raise a promotion request and approve a high-impact action, so your risk team can see the record each step leaves.

Book a Warsha demo

We only use your details to reply to you. Arabic or English, as you prefer.