AI Auditing Masterclass: Build AI Audit Expertise

Explore how AI evolves from automation to complex decision making, and how auditors must audit intelligence itself, not just code, while assessing governance and compliance risks.

Understand four ai system types, from narrow ai to general ai, and audit implications; examine machine learning categories: supervised, unsupervised, and reinforcement, plus deep learning, generative ai, large language models.

Explore supervised learning through classification and regression models, training and validation processes, and performance metrics, highlighting bias risks and auditing considerations for real-world applications like fraud detection and pricing.

Explore unsupervised learning to reveal hidden data patterns through clustering, dimensionality reduction, pattern recognition, and anomaly detection, addressing auditing challenges and ensuring fair, accurate, and compliant decision making.

Explore reinforcement learning fundamentals—agent–environment interactions, reward systems, and policy optimization—and see how real-world applications like data center cooling, ride matching, and content recommendations reveal system behavior and auditing implications.

Discover natural language processing techniques that let machines understand, interpret, and generate language. Explore text processing, sentiment analysis, and machine translation used in audits and business.

Explore how computer vision enables image recognition, object detection, and facial recognition, with applications in healthcare, vehicles, manufacturing, and retail, while examining accuracy, bias, privacy, and safety considerations for auditors.

Learn how data pipelines and ETL, model training infrastructure, and deployment choices—cloud, on premise, and edge—shape AI reliability, privacy, and performance for auditors.

Define technical ai terminology to distinguish algorithms from models and explain training, validation, and testing for auditors. Examine hyperparameters, feature engineering, and overfitting and underfitting to assess real-world model performance.

Explore the AI lifecycle from problem definition to production, covering deployment, monitoring, drift and degradation, A/B testing, governance, and audit-ready controls to sustain performance.

Explore traditional IT audit foundations built on Cobit and ITIL governance, control objectives, testing procedures, and evidence-based documentation, and learn how to adapt these frameworks for AI auditing.

Explore traditional IT risk assessment foundations and expand into AI-specific threats like algorithmic bias and model drift, with controls, compliance testing, audit reporting, and continuous follow-up.

Audit AI systems through four characteristics: non-deterministic behavior, continuous learning, black box versus explainable AI, and dynamic decision making; develop new testing methods, guardrails, and ensure regulatory requirements are met.

Assess model performance beyond output accuracy by evaluating prediction quality and robustness. Detect bias with fairness testing, ensure data quality and lineage, and verify algorithmic transparency for trusted AI decisions.

Develop the ai auditor mindset by balancing technical depth, governance, and ethical awareness, applying critical thinking and supervised and unsupervised learning basics to audit machine learning systems responsibly.

Explore a robust policy framework for AI governance, covering algorithmic accountability, data governance, ethical use, model management, deployment protocols, and ongoing monitoring to prevent bias and ensure transparency.

Develop AI governance literacy across the organization by delivering executive education, technical training, and end user awareness to manage risks, ethics, and value from AI initiatives.

Measure success across AI programs with performance, compliance, ethical impact, and operational metrics. Align AI outcomes with business impact, governance, risk assessments, and responsible bias and fairness considerations.

Explore ISO/IEC 42001's AI management system framework, governance, life-cycle management, risk, phased implementation, and audits and certification integrated with existing quality systems.

Explore the NIST AI risk management framework, detailing its four core functions: govern, map, measure, manage, plus risk assessment methodologies, implementation guidance, and continuous measurement for AI risk management.

Explore IEEE AI standards for privacy engineering, explainable AI, and bias management, with privacy by design, risk assessment, and integrated governance to audit trusted, transparent AI systems.

Explore the EU AI act’s risk-based classification system with four categories, compliance requirements, and conformity assessments, from minimal to prohibited practices, with real-world audit guidance.

Navigate a complex landscape of regional and national AI regulations, including US sector-specific rules, FDA and FTC guidance, and Asia's governance models, and cross-border compliance.

Explore four dimensions of human in the loop systems—oversight, decision authority, intervention protocols, and performance monitoring—to foster meaningful human AI collaboration (air traffic control, radiology, warehouse robotics).

Examine accountability frameworks for AI systems across responsibility assignment, liability, audit trails, and escalation procedures to ensure clear ownership and continuous improvement throughout the AI lifecycle.