Reproduce “Three Kinds of AI Ethics” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Emanuele Ratti
Emanuele Ratti proposes a taxonomy dividing AI ethics into three distinct types: Ethics and AI (EaA), Ethics in AI (EiA), and Ethics of AI (EoA). This framework clarifies the differing questions, methodologies, and necessary expertise for each category, offering a structured approach to the field and addressing common misdirected criticisms.
Key insights
- The relationship between 'AI' and 'ethics' can be systematically categorized into three types, each leading to fundamentally different research questions, goals, and required expertises.
- Ethics and AI (EaA) involves external evaluation of AI functions, Ethics in AI (EiA) focuses on embedding ethics into AI's internal design, and Ethics of AI (EoA) examines the mutual shaping of AI and its socio-technical context.
- Many criticisms directed at AI ethics are often misplaced, stemming from judging one kind of AI ethics by the standards or goals of another.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.