Interoceptive Divergence in Aesthetic Evaluation and Implications for Human-AI Alignment

TL;DR

This study compares human and AI aesthetic evaluations, revealing similarities in emotion and feature prioritization but notable differences in emotional distribution and bodily sensation correlations, highlighting challenges in AI alignment.

Contribution

It introduces a method to directly compare human and AI aesthetic responses using questionnaires, revealing key divergences in interoceptive aspects of aesthetic evaluation.

Findings

Humans and AI show similar patterns in beauty-emotion correlations.

AI approximates average human aesthetic tendencies but differs in bodily sensation responses.

Limitations in AI interoceptive processing may stem from training data or alignment methods.

Abstract

Artificial intelligence (AI), exemplified by large language models (LLMs), is rapidly approaching and in some cases surpassing human performance across a wide range of cognitive tasks. However, human nature is not limited to intelligence alone; it also encompasses sensibility, including the capacity to perceive and experience beauty in visual scenes. This raises a fundamental question: how humans and AI systems converge or diverge in such aesthetic experiences. Aesthetic evaluation depends not only on objective properties of images but also on internal processes within the observer. As part of ongoing efforts in AI alignment, building upon prior human studies that have examined the relationship between beauty ratings, bodily sensations, and emotions, we adopt a comparable set of questionnaire items and present them to LLMs, enabling a direct comparison between human and AI responses. Our comparative analyses revealed that, while humans and AI exhibited broadly similar patterns in the correlations between beauty ratings and emotions, as well as in the image features they prioritized, notable divergences emerged in both the distribution of emotional responses and the relationship between beauty ratings and bodily sensations. These findings suggest that state-of-the-art LLMs, trained on large-scale textual data, can approximate average human tendencies in aesthetic evaluation to a certain extent. However, they also indicate limitations, particularly in relation to interoceptive aspects, which may reflect insufficient representation in training data or unintended consequences of alignment processes. These findings highlight key challenges for AI alignment and suggest important directions for developing AI systems with human-like aesthetic processing.