Case Study PDFs
The Princeton Dialogues on AI and Ethics was launched in 2017. As of July, 2020, two final case studies have been released, and due to changes in personnel, the initiative has been put on hold indefinitely.
Below are a set of fictional case studies that are designed to prompt reflection and discussion about issues at the intersection of AI and Ethics. These case studies were developed out of an interdisciplinary workshop series at Princeton University that began in 2017-18. They are the product of a research collaboration between the University Center for Human Values (UCHV) and the Center for Information Technology Policy (CITP) at Princeton. Click the title of each case study to download the full document.
Case Study 1: Automated Healthcare App
Issues:
Foundations of legitimacy, Paternalism, Transparency
Censorship, Inequality
Case Study 2: Dynamic Sound Identification
Issues:
Rights, Representational harms, Neutrality, Downstream responsibility
Case Study 3: Optimizing Schools
Issues:
Privacy, Autonomy, Consequentialism, Rhetoric
Case Study 4: Law Enforcement Chatbots
Issues:
Automation, Research ethics, Sovereignty
Case Study 7: Autonomous Vehicles
Issues:
The Trolley Problem,
Deontological ethics,
Utilitarianism,
Collective responsibility
Case Study 5: Hiring By Machine
Issues:
Fairness, Irreconcilability, Diversity, Capabilities,
Contextual integrity
Case Study 8: Legal E-Discover System
Issues:
Accuracy,
Competence,
Job displacement
Case Study 6: Public Sector Data Analytics
Issues:
Democracy, Secrecy, Inequality, Fallibility,
Determinism