Categorical AI phenomenology: A first-person approach

This paper proposes a new method called 'categorical AI phenomenology' that applies phenomenological principles to study first-person experiences of interacting with artificial intelligence. The authors argue that existing approaches to understanding AI experience are inadequate because they rely on third-person behavioral or computational descriptions. Instead, they contend that a categorical phenomenological framework, drawing on Husserlian and Merleau-Pontian traditions, can systematically capture the essential structures of human-AI interaction as lived experience. The paper outlines how this method might be implemented through structured first-person reports and phenomenological reduction, aiming to provide a rigorous qualitative methodology for studying subjective encounters with AI systems.

Study at a glance

Abstract

Categorical AI phenomenology: A first-person approach