Document and Evaluate AI Ethics

Overview

Document and Evaluate AI Ethics is an intermediate course that equips engineers, auditors, and AI practitioners with the concrete skills to move from ethical principles to engineering practice. You will learn to create comprehensive model cards that document a system's intended use, dataset origins, performance metrics, and limitations, ensuring every stakeholder understands what the system does and where it might fail. Next, you will master the process of conducting systematic ethics audits, using established frameworks to evaluate AI systems for bias, assess compliance, and propose actionable mitigation strategies. Through hands-on labs and analyses of real-world case studies—from the failure of Microsoft’s Tay to the internal audits at AstraZeneca—you will leave with the ability to produce professional audit reports and documentation that build trust and ensure your AI systems are deployed responsibly.

Syllabus

Create Comprehensive Model Documentation

Have you ever released an AI feature only to realize no one could clearly explain what it does or where its data came from? That gap begins with missing documentation. This module shows how structured model cards build trust and accountability. You will capture a model’s purpose, dataset origins, fairness metrics, and limits, creating documentation that speaks to both engineers and stakeholders and keeps AI systems transparent.

Conduct AI Ethics Audits

Have you ever seen an AI output that felt biased or off and wondered what caused it? Ethical issues often hide in oversight, not code. This module shows how to identify and address those risks before they escalate by applying structured audit frameworks to evaluate fairness and compliance in AI systems.