Perceptions of AI-generated and co-created music among university students and social media users

Abstract

This study explores how listeners perceive and interpret music created by or with artificial intelligence (AI). Using a custom-designed survey, 57 university students and social media users listened to a series of 12 music tracks (composed by humans, co-created with AI, or fully AI-generated) and attempted to identify the origin of each. Participants also responded to questions about their attitudes toward AI in music. Results revealed that co-created tracks were the most difficult to identify correctly, while AI-only and human-generated tracks had modestly higher identification rates. Musicians and non-musicians showed similar accuracy but differed in confidence and willingness to engage with AI-assisted music. Most participants expressed support for labeling AI-generated music and were cautiously open to AI-human creative collaboration, though concerns remained regarding authenticity and emotional depth. These findings offer insight into public readiness for AI in the arts and suggest pathways for ethical design, media literacy, and future research in music technology.