Born in the wrong era: rethinking human-AI relations through the lens of anemoia

Abstract

As large language models (LLMs) become increasingly integrated into everyday communication, they largely transform how users access information and reshape social dynamics. This study explores how LLM users 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 1625 Reddit comments, this research reveals that AI-related discourses frequently reflect discomfort, emotional fatigue, and yearning for simpler, more human-centered eras. Using Grounded Theory (GT) coding procedures, the study identifies recurring discursive patterns consistent with AI-triggered anemoia . The findings suggest that anemoia provides a basis for understanding why some users feel emotionally ambivalent toward AI. By conceptualizing AI-triggered anemoia as a recurring interpretive pattern in user discourse, the study offers implications for emotionally informed AI design and future research on digital well-being.