Who Gets Heard? Rethinking Fairness in AI for Music Systems

TL;DR

This paper highlights cultural and genre biases in AI music systems, emphasizing their impact on marginalized traditions and proposing recommendations to improve fairness at various system levels.

Contribution

It identifies biases in music-AI systems affecting marginalized cultures and offers specific recommendations for dataset, model, and interface improvements.

Findings

Biases distort representations of marginalized music traditions

Inauthentic outputs undermine trust among creators and listeners

Recommendations aim to mitigate cultural biases in AI music systems

Abstract

In recent years, the music research community has examined risks of AI models for music, with generative AI models in particular, raised concerns about copyright, deepfakes, and transparency. In our work, we raise concerns about cultural and genre biases in AI for music systems (music-AI systems) which affect stakeholders including creators, distributors, and listeners shaping representation in AI for music. These biases can misrepresent marginalized traditions, especially from the Global South, producing inauthentic outputs (e.g., distorted ragas) that reduces creators' trust on these systems. Such harms risk reinforcing biases, limiting creativity, and contributing to cultural erasure. To address this, we offer recommendations at dataset, model and interface level in music-AI systems.