Many professionals asked to oversee AI understand the principles but have never filled in an impact assessment, read a fairness report or checked a model's documentation against a standard. That gap matters. Regulators, auditors and certification bodies look for records and test results, not intentions, and both ISO/IEC 42001 and the EU AI Act expect evidence that risks were assessed and controls actually operated.

The Practical AI Governance, Responsible AI & ISO 42001 Training Course is organised as a sequence of build sessions. In the first week participants set up the core governance records for a sample organisation: inventory, policy, scope, risk register, impact assessment and control selection. In the second week they work with the technical evidence behind those records, testing a credit-scoring model for bias, generating explanations, documenting models and datasets, configuring drift reports, red teaming a generative AI assistant and finally assembling everything into an audit-ready evidence file.

Every day follows the same rhythm of short demonstration, guided build and peer review. Templates are supplied in spreadsheet and document form, and the Python notebooks run in a browser with prepared datasets, so participants concentrate on interpreting results rather than writing code.

Working through the hands-on sessions, participants will be able to:

• Build and maintain an AI system inventory that records ownership, purpose, data sources and risk tier.
• Complete an AI risk assessment and treatment plan using a scoring matrix aligned with ISO/IEC 23894.
• Carry out an AI system impact assessment and record its findings for decision makers.
• Draft a Statement of Applicability that justifies the inclusion or exclusion of ISO/IEC 42001 Annex A controls.
• Measure fairness across groups with Fairlearn and interpret the results for non-technical colleagues.
• Explain model predictions with SHAP and LIME and record known limits in a model card.
• Configure drift and performance monitoring and write the alert and response rules.
• Assemble an audit-ready evidence file and test it in a mock internal audit.

The course is built for hands-on practitioners who will produce or review governance records, such as:

• Risk officers and compliance analysts asked to review AI use cases
• Data protection officers and privacy analysts
• Data analysts, data scientists and machine learning engineers
• IT auditors and internal audit staff assessing AI controls
• Product managers and business analysts introducing AI features
• Information security analysts and ISO/IEC 27001 implementers
• Quality and documentation officers supporting certification projects
• ICT and digital officers in ministries, local authorities and public agencies
• Monitoring, evaluation and data staff in NGOs using predictive or generative tools

Participants go home with a working toolkit, including:

• A populated AI inventory workbook that can be reused in their own organisation.
• A risk register and treatment plan with scored, justified entries.
• A completed impact assessment template with worked examples of findings.
• A draft Statement of Applicability for the ISO/IEC 42001 Annex A controls.
• Reusable Python notebooks for fairness testing and model explanation.
• A model card and datasheet written for a sample credit-scoring model.
• A generative AI red-teaming log and acceptable use checklist.
• An indexed evidence file ready for internal audit review.

Every session is built around doing the work: the trainer demonstrates each step, then participants repeat it on their own files. The course combines:

• Guided builds using supplied spreadsheet and document templates
• Browser-based Python notebooks with prepared datasets, usable without prior coding
• A single sample organisation carried through every exercise
• Pair work in which one participant builds and another reviews
• Red-teaming and incident tabletop sessions with live scenarios
• A mock internal audit of each participant's evidence file

Completing the hands-on labs, attending every session and submitting the final evidence file earns each participant a CPD-accredited Certificate of Completion issued by Vision Reach Global Consultancy.

The certificate attests to the training and practical work completed. It does not certify an organisation to ISO/IEC 42001 or qualify the holder as a lead auditor, and participants keep the full toolkit they built.