Auditing AI Systems and Algorithms
Equip auditors to plan and perform audits of AI systems themselves: understanding architecture and data provenance, using model documentation, setting objectives and evidence requirements, testing accuracy, robustness, explainability, bias and drift, and reporting findings, limitations and assurance conclusions accurately.
Upcoming course. LMS delivery and enrolment are not open. Release dates will be confirmed.
Express interest →
Planned learning outcomes
- Describe an AI system's architecture and trace data provenance and lineage
- Evaluate model documentation, inventories and change records as audit evidence
- Set audit objectives, criteria, scope, methodology and evidence requirements for an AI system audit
- Plan and interpret tests of accuracy, robustness and explainability
- Plan and interpret bias and drift tests, recognising their limitations
- Report findings, limitations and the level of assurance accurately