Andrea Cuadr
From ELIZA to Alexa, Conversational Agents (CAs) have been delib erately designed to elicit or project empathy. Although empathy can help technology better serve human needs, it can also be deceptive and potentially exploitative. In this work, we characterize empathy in interactions with CAs, highlighting the importance of distin guishing evocations of empathy between two humans from ones between a human and a CA. To this end, we systematically prompt CAs backed by large language models (LLMs) to display empathy while conversing with, or about, 65 distinct human identities, and also compare how diferent LLMs display or model empathy. We fnd that CAs make value judgments about certain identities, and can be encouraging of identities related to harmful ideologies (e.g., Nazism and xenophobia). Moreover, a computational approach to understanding empathy reveals that despite their ability to display empathy, CAs do poorly when interpreting and exploring a user’s experience, contrasting with their human counterparts. CCS CONCEPTS • Human-centered computing → HCI theory, concepts and models; Empirical studies in HCI.