From the course: Introduction to Auditing AI Systems
Unlock the full course today
Join today to access over 24,600 courses taught by industry experts.
Continuous audits
From the course: Introduction to Auditing AI Systems
Continuous audits
- [Instructor] Continuous AI auditing for compliance and bias mitigation is an essential process to make sure models remain safe for use. It's not enough to simply check a model for compliance or fairness once. As AI tools quickly learn from their environments, we should continuously monitor their performance and conduct audits at meaningful intervals. Often, continuous audits can be an extension of prior audits to better cover the scope of models in an AI system. This is pretty common as new AI enabled features are added to products and need to undergo auditing. In some cases, further audits attempt to expand the depth of prior audits. If a past audit had to be reduced in scope to only audit the data, a further audit might be a good time to audit the models or conduct a comprehensive system audit. Organization should work with users, data subjects, and regulators to identify what an appropriate audit frequency would be.…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.