DE

AIMS · Governance · certification readiness

ISO/IEC 42001:2023: Why is the international standard for AI governance not enough on its own?

By Ing. Mag. Patrick Eiler

ISO/IEC 42001 creates a robust framework for an AI management system. EU AI Act-specific duties such as role allocation, transparency, FRIA and market surveillance still need to be assessed separately.

What is ISO/IEC 42001?

ISO/IEC 42001:2023 is the first international standard for an AI Management System (AIMS). It describes how organisations can develop, provide or use AI responsibly, transparently and with evidence.

For institutions with an existing ISO 27001 or ISO 9001 management system, this is helpful because ISO/IEC 42001 follows the familiar management-system logic: context, leadership, planning, support, operation, performance evaluation and improvement.

What does ISO/IEC 42001 contribute to the EU AI Act?

The standard supports governance, risk and quality management, competence, documentation, supplier control and continuous improvement. It is therefore a strong framework for anchoring EU AI Act compliance organisationally.

  • Risk management and documented assessment processes
  • Data governance and quality requirements
  • Human oversight and responsibilities
  • Competence, awareness and training
  • Supplier and lifecycle management

Where does ISO/IEC 42001 reach its limits?

Five topics need separate checks: role allocation, CE marking and EU database, FRIA, incident response and labour law.

AreaEU AI Act requiresAdditionally required
Role allocationDistinction between provider, deployer, importer and distributorLegal and process assessment for each system and use case
CE / EU databaseProvider duties for high-risk AIConformity assessment, technical documentation and registration where a provider role exists
FRIAFundamental Rights Impact Assessment for certain high-risk usesSpecific impact assessment for credit and life/health insurance and possibly other constellations
Incident responseReporting obligations under the EU AI Act, GDPR and DORACoordinated three-regime process with clear deadlines and escalation paths
Labour lawInformation and possibly co-determination for HR AIInvolvement of employee representatives and review of Austrian labour constitution requirements

Which AI Act duties does ISO/IEC 42001 not fully cover?

Five formal duties: conformity assessment (Art. 43), CE marking (Art. 48), EU database registration (Art. 49), post-market monitoring (Art. 72) and incident response (Art. 73). ISO/IEC 42001 is a strong management-system framework, but certification does not automatically replace these duties. The gaps should be documented explicitly.

GapEU AI ActAdditional clarification
Conformity assessmentArt. 43Formal procedure for each provider role and high-risk system.
CE markingArt. 48Relevant where provider duties apply; process and evidence must be handled separately.
EU database registrationArt. 49Registration duties for certain high-risk systems must be assessed separately.
Post-market monitoringArt. 72AI-Act-specific monitoring and feedback with provider information must be defined.
Incident responseArt. 73, GDPR, DORA, NISG 2026Reporting cascade with different authority routes and timelines must be mapped.

Which harmonised EU standards exist for the AI Act?

CEN/CENELEC are developing harmonised standards for the EU AI Act. The first standard for AI Act purposes, EN 18286:2026 (quality management system for AI Act purposes, in particular Art. 17 for providers of high-risk AI), was announced as published by CEN/CENELEC on 31 July 2026. Publication of the standard and citation of its reference in the Official Journal of the EU are two separate steps: only the Official Journal citation triggers a presumption of conformity. For EN 18286 it was published by Implementing Decision (EU) 2026/2274 on 14 October 2026; the presumption covers the Art. 17 requirements addressed by the standard. ISO/IEC 42001 is a strong foundation, but it does not automatically replace this assessment.

CSRD / ESRS: Can ISO 42001 serve as an evidence framework?

Yes, as an organisational framework. If AI bias or discriminatory AI decisions are classified as material sustainability impacts or governance risks, CSRD / ESRS reporting needs robust evidence. ISO/IEC 42001 can provide the organisational framework: policies, roles, risk assessments, controls, monitoring, internal audits and management reviews. It does not replace the ESRS reporting obligation, but it makes evidence much easier to structure. This presupposes that the institution is subject to reporting at all: Directive (EU) 2026/470 has limited the CSRD scope to EU companies with more than 1,000 employees and more than EUR 450 million net turnover; national transposition, transition years and group-level assessment must also be checked. The impact or risk must also be material – not every bias incident leads to an ESRS disclosure.

What an ISO/IEC 42001 implementation means

Implementation is not a pure IT project. It affects the board, compliance, legal, IT, HR, procurement and business units. Core elements are scope, AI inventory, risk and control model, roles, evidence and regular management reviews.

Ing. Mag. Patrick Eiler

About the author

Ing. Mag. Patrick Eiler

Patrick Eiler advises banks and insurers in Vienna on regulatory reporting, data architecture and AI compliance. More than twenty years in banking, insurance and financial services; speaker at the Hypo-Verband (expert session of 23 September 2026).

Where does your institution stand on the path to ISO/IEC 42001?

A gap analysis shows which building blocks already exist and where a realistic roadmap should begin.

Request gap analysis
Notice: This page is for general information only and does not constitute certification or legal advice. Requirements may become more specific through harmonised standards and regulatory interpretation. Status: October 2026.