CSAI: Case Studies in AI Ethics (CSAI)

Welcome to Case Studies in AI Ethics (CSAI)

Learning Outcomes

On successful completion of this course, you should be able to: 

  1. understand data ethics and arising issues (e.g. bias, fairness, privacy) in AI systems
  2. explain and provide examples of how AI systems can play a critical role in decision making
  3. analyse case studies to identify and mitigate potential risks considering legal, social, ethical or professional issues
  4. apply ethical methodologies in the design of responsible AI systems
Course Outline

This course follows the delivery and assessment of Case Studies in AI Ethics (CSAI) (INFR11206) exactly. Undergraduate students must register for this course, while MSc students must register for INFR11206 instead.

In this course, we will discuss the following topics:

Data Ethics:
- Deployed AI technologies
- Ethical and social issues arising with data

Fairness, Accountability and Transparency:
- Overview of the definitions
- Types of bias
- Explainability

Privacy:
- Arising issues (e.g. surveillance, usability vs privacy trade-off)
- State of the art: ML approaches, Agent-based approaches

Towards implementing ethical tools:
- Implementing AI Ethics 
- Ethics guidelines for Trustworthy AI (e.g. European Commission), AI Auditing guidelines (e.g. ICO) 
- Applied Ethics (e.g. IEEE Ethics in Action, Markkula Centre's Ethics Toolkit)

The students will be expected to prepare for the lectures by reading papers, news; or watching videos. Some lectures will include case studies where students will discuss the ethical issues in small discussion groups for 15 minutes; and report back their findings.

License
All rights reserved The University of Edinburgh