New Study Reveals Shocking Gaps in AI Empathy

Conversational agents (CAs) like Alexa and Siri are designed to answer questions, offer suggestions, and even display empathy. However, new research indicates that they fall short compared to humans in interpreting and exploring a user’s experience.

CAs are powered by large language models (LLMs) that ingest massive amounts of human-produced data, and thus can be prone to the same biases as the humans from which the information comes.

Researchers from Cornell University, Olin College, and Stanford University tested this theory by prompting CAs to display empathy while conversing with or about 65 distinct human identities.