PH440      Half Unit
The Ethics of Data and AI

This information is for the 2025/26 session.

Course Convenor

Availability

This course introduces you to the core philosophy of science, philosophy of mind, and ethics concepts needed to build better technology and reason about its impact on the economy, civil society, and government.

Some questions that the course might consider include:

• What is intelligence, and how does it vary between types of agents (human, animal, artificial)? What are the normative assumptions behind research in intelligence?
• What is data, and how can we design more ethical data governance regimes?
• Can technology be racist? If so, what are promising strategies for promoting fairness mitigating algorithmic bias?
• Can we understand black box AI and explain its outputs? Why is it morally important that we do so?
• How can we embed human values into AI systems?

15 hours of seminars and 10 hours of lectures in the Winter Term.

This course has a reading week in Week 6 of Winter Term.

Students are invited to submit a mock exam, answering one question from a list of sample exam questions provided in advance. This should be written as a practice exam: completed in one hour, in a single sitting, and without the use of notes. The essay should be no longer than 1,000 words.

• Gabriel, “Towards a Theory of Justice for Artificial Intelligence”, Daedalus
• Friedman, Kahn, and Borning, “Value Sensitive Design and Information Systems”
• Serpico “What kind of kind is intelligence?”
• Henry Shevlin, Karina Vold, Matthew Crosby & Marta Halina, “The limits of machine intelligence”
• Halina, “Insightful artificial intelligence”
• Alexandrova and Fabian, “Democratizing Measurement: Or Why Thick Concepts Call for Coproduction”
• Northcott, “Big Data and Prediction: Four Case Studies”
• Simons and Alvarado, “Can we trust Big Data? Applying philosophy of science to software”
• Viljoen, “A Relational Theory of Data Governance”
• Johnson, “Are Algorithms Value Free?”
• Munton, “Beyond accuracy: Epistemic flaws with statistical generalizations.”
• Barocas, Hardt, and Narayanan, “Fairness and Machine Learning: Limitations and Opportunities” [selections]

Exam (70%), duration: 120 Minutes in the Spring exam period

Project (30%)

The group project will be assessed through a class presentation worth 30% of the final mark. The presentation will include visual materials (such as slides or a poster) along with a verbal explanation. Students who are unable to present in class due to Disability and Wellbeing reasons will instead submit an individual 1,000-word write-up (also worth 30%).