Qualitative analysis reveals how large language models affect emotional responses in users, suggesting deeper social implications.
Abstract As large language models [LLMs] become increasingly integrated into everyday communication, they not only transform how users access information but also reshape emotional and social dynamics. This study explores how individuals emotionally respond to the growing presence of AI, with a particular focus on anemoia – a nostalgic longing for a past barely or never personally experienced. Drawing on a qualitative content analysis of 1.625 Reddit comments, this research reveals that user discourse frequently reflects discomfort, emotional fatigue, and yearning for simpler, more human-centered eras. Using a Grounded Theory [GT] approach, the study identifies core emotional themes, including alienation, perceived loss of authenticity, and digitally mediated loneliness, and demonstrates how these are expressed on a spectrum of anemoia. The findings position anemoia as a useful interpretive lens for understanding emotional ambivalence toward AI, revealing how imagined pasts influence user attitudes. By introducing the concept of AI-triggered anemoia, the study contributes to emerging debates on affective responses to AI and calls for more psychologically sensitive approaches to AI development.
No takes yet. Share an insight, caveat, or question.
Márk Miskolczi (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: