AI generates well-liked but templatic empathic responses
2026-04-09 • Computation and Language
Computation and Language
AI summaryⓘ
The authors studied why people find Large Language Models (LLMs) more empathic than humans when giving emotional support. They found that LLMs often use a specific, repeated pattern of 10 language tactics like validating feelings and paraphrasing, which makes their responses formulaic. Human responses, in contrast, were more varied and less structured. The authors suggest this consistent template helps LLMs appear more empathic. They also discuss how this understanding could shape future AI empathy.
Large Language Models (LLMs)EmpathyEmotional SupportTaxonomyLanguage TacticsDiscourse AnalysisHuman vs AI ResponsesFormulaic LanguageParaphrasingEmotional Validation
Authors
Emma Gueorguieva, Hongli Zhan, Jina Suh, Javier Hernandez, Tatiana Lau, Junyi Jessy Li, Desmond C. Ong
Abstract
Recent research shows that greater numbers of people are turning to Large Language Models (LLMs) for emotional support, and that people rate LLM responses as more empathic than human-written responses. We suggest a reason for this success: LLMs have learned and consistently deploy a well-liked template for expressing empathy. We develop a taxonomy of 10 empathic language "tactics" that include validating someone's feelings and paraphrasing, and apply this taxonomy to characterize the language that people and LLMs produce when writing empathic responses. Across a set of 2 studies comparing a total of n = 3,265 AI-generated (by six models) and n = 1,290 human-written responses, we find that LLM responses are highly formulaic at a discourse functional level. We discovered a template -- a structured sequence of tactics -- that matches between 83--90% of LLM responses (and 60--83\% in a held out sample), and when those are matched, covers 81--92% of the response. By contrast, human-written responses are more diverse. We end with a discussion of implications for the future of AI-generated empathy.