Generative AI, Ethics, and Inequity in Popular Music
Simulated Voices, Real Consequences responds to a critical moment in which generative artificial intelligence (GenAI) is being rapidly integrated into popular music in advance of clear legal, regulatory, or industry-wide ethical frameworks. The project is organized around focused case studies of AI-generated and AI-assisted artists – Xania Monet, Breaking Rust, and Randy Travis – through which it examines how GenAI is being positioned as a cultural and commercial actor within the music industry. These case studies combine close musical analysis of AI-generated recordings with analysis of media, industry, and artists’ discourse, grounding ethical questions in both recorded sound and public debate. Complementing this work, the project incorporates semi-structured interviews with industry stakeholders, including representatives of musicians’ unions, songwriters, producers, researchers, and journalists, to address broader contextual, practical, and policy-relevant issues that extend beyond individual cases. Rather than offering retrospective critique, the project intervenes early, contributing conceptual clarity and ethical reflection at a moment when norms around consent, authorship, control, compensation, and accountability are still being negotiated.
The project is guided by a feminist ethic of care, which foregrounds relationality, responsibility, and the uneven distribution of risk and benefit within socio-technical systems. Building on ethical principles articulated by Joy Buolamwini – consent, credit, control, and compensation – the project foregrounds a fifth concern: consequences. This extension situates GenAI within a longer historical lineage of extraction, exclusion, and unequal power relations in the popular music industry, asking not only who authorizes and benefits from GenAI systems, but who bears their downstream cultural, social, and economic effects. The framework also recognizes the potential of GenAI as an assistive technology, particularly for artists with disabilities, while distinguishing such uses from extractive practices that simulate or displace marginalized creators without accountability.
This project marks a departure in my research, involving significant changes in focus, theoretical orientation, and method. Moving beyond prior analyses of representation and inequity in the country music industry, it examines the ethical, cultural, and industry implications of generative AI through qualitative, ethics-informed humanities approaches. In the absence of established governance mechanisms, the project’s significance lies in its capacity to inform how norms are formed before they become entrenched, supporting more accountable and equitable approaches to GenAI in music.
Because these debates extend beyond academic contexts, knowledge mobilization is a core component of the project. Alongside scholarly publications, it will generate public-facing outputs – including a podcast series and public essays – aimed at artists, educators, policymakers, and the public. As a Canadian-based project engaging a transnational industry, it will contribute humanities-grounded perspectives to Canadian cultural and policy conversations concerning copyright, consent, and the governance of creative training datasets.
Simulated Voices, Real Consequences contributes a humanities-based ethical framework for analyzing GenAI in popular music that integrates musical analysis, industry discourse, and stakeholder perspectives, offering a model for future research on creative AI beyond music. In doing so, it positions Canadian humanities scholarship within global debates on creative AI, contributing critical perspectives on copyright, consent, and cultural labour in a transnational industry.
Stay tuned for more on this project!
This project was approved by the Research Ethics Board of the University of Ottawa on 27 July 2026 (approval number S-07-26-13107). Written informed consent was obtained from all participants prior to data collection.
This research is supported by an Insight Development Grant from the Social Sciences and Humanities Research Council of Canada.
