AI150: AI Lead Auditor ISO/IEC 42001:2023 (DEKRA™ AILA)

Agenda:

• Technical fundamentals & AI management system

  • Introduction to AI systems

  • Types of AI systems and their applications

  • Basic AI architecture and mode of operation

  • Current developments and trends

  • Purpose, structure and scope of ISO/IEC 42001

  • Legal and regulatory requirements (e.g. GDPR, AI Act, Data (Governance) Act) in the AI context
    
    

• Audit fundamentals & auditor competence

  • Audit principles according to ISO 19011

  • Standards for lead auditors

  • Interaction of ISO/IEC 42001 and ISO/IEC 42006 (ISO 17021)
    
    

• Audit planning & preparation

  • The certification process according to ISO 17021

  • Structure of an audit programme for AI management systems

  • Risk-based audit planning (objectives, scope, criteria)

  • Integration of AI assessment aspects into audit planning

  • Preparation of audit checklists and document reviews

  • Roles, responsibilities and professional audit ethics

  • Leading an audit team: role allocation, delegation, time and resource management, communication
    
    

• Audit execution & interview strategies

  • Execution of Stage 2 audits (on-site / remote)

  • Interview strategies in AI audits: open, in-depth and targeted questioning techniques

  • Dealing with subject matter experts and management as well as with resistance

  • Evidence collection related to AI: training data, model validation, bias detection

  • Evaluation of AI-specific risks and nonconformities
    
    

• AI-technical test methods & audit practices

  • Ethical assessment of AI systems

  • Technical evaluation: training data, data pipelines, model performance, bias detection, fairness, security and robustness

  • Quality assurance techniques and audit tests

  • Integration of specific test procedures: VDE SPEC 90012, TÜV AI.AM Assessment Matrix
    
    

• Audit closure, reporting & follow-up

  • Evaluation and consolidation of audit findings

  • Dealing with difficult audit situations

  • Closing meetings with audited functions

  • Preparation of meaningful audit reports

  • Handling of corrective and improvement measures

  • Ensuring traceability and audit trails in AI audits
    
    

• Practice, case studies & examination

  • Audit simulations including team leadership and interview exercises

  • Case studies from real AI audit contexts

  • Review of key audit principles and AI specifics

  • Qualifying DEKRA™ AI Lead Auditor examination (including AI audit elements)
    
    

• Didactic focus

  • Practice-oriented audit role plays and realistic scenarios

  • Interactive group work, reflection and peer feedback

  • Integration of technical AI assessment aspects into classical audit methodology

  • Teaching soft Skills for team leadership and communication

Objectives:

• gain an in-depth understanding of the requirements and structure of an AI Management System (AIMS) in accordance with ISO/IEC 42001,
• gain an in-depth understanding of the certification process according to ISO 17021 in combination with ISO/IEC 42006,
• translate the relevant legal frameworks and standards (e.g. EU AI Act, GDPR) into audit contexts,
• plan audit programmes on a risk-based basis, including audit objectives, scope and criteria,
• prepare and conduct audits in a structured manner, with a particular focus on AI-related requirements and risks,
• lead audit teams in a professional manner and apply audit methods, including effective interview and questioning techniques,
• assess technical and qualitative aspects of AI systems (e.g. data, model validation, bias, security) in the audit,
• correctly identify, document and follow up nonconformities and audit findings,
• prepare audit reports in a clear and transparent manner and professionally facilitate closing meetings,
• acquire practice-oriented knowledge of audit execution, the application of relevant test procedures and quality assurance measures, combined with the ability to apply ethical, security and regulatory requirements in their own audit contexts.
  
  
  
  
  
  

Target audience:

• want to introduce, operate or improve a formal AI Management System (AIMS) in accordance with ISO/IEC 42001
• as auditors wish to deepen their knowledge and skills in auditing management systems with a specific focus on artificial intelligence
• as consultants or specialists want to advise companies on ISO/IEC 42001 certifications and AI audit processes
• work in the areas of quality, risk or compliance management and wish to expand their competencies to include AI audit methods
• are required to assess and evaluate technical, organizational or regulatory aspects of AI systems as part of audits

Prerequisites:

• Prior attendance of the AI120 AIMS Implementation according to ISO/IEC 42001:2023 or experience with management systems as well as prior basic engagement with ISO/IEC 42001:2023
• Prior attendance of the SC135 Internal Auditor or comparable audit knowledge

Description:

The course AI150 AI Lead Auditor ISO/IEC 42001:2023 (DEKRA™ AILA) provides practical knowledge for auditing AI Management Systems (AIMS) according to ISO/IEC 42001:2023. Participants gain in-depth understanding of the requirements and structure of an AIMS as well as the certification process according to ISO 17021 in combination with ISO/IEC 42006.

They learn to plan audit programs risk-based, to prepare and conduct audits in a structured manner, and to lead audit teams professionally – with particular focus on AI-related requirements and risks. Technical aspects such as training data, data pipelines, model validation, bias detection, fairness as well as security and robustness of AI systems are embedded into audit practice.

Legal and regulatory frameworks – including the EU AI Act and GDPR – are transferred into audit contexts in a practical manner. Practice-oriented role plays, realistic audit scenarios as well as interactive group work with peer feedback prepare specifically for the DEKRA™ AI Lead Auditor certification.

Other Info:

The DEKRA™ certification exam is conducted fully online. For participation in the online exam, each participant requires their own notebook. The exam can be taken following the course; the exact exam period will be communicated after schedule confirmation.