Reproduce “Avoiding an AI-imposed Taylor's Version of all music history” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Nick Collins, Mick Grierson
This research explores the "Taylor Swift problem," a speculative scenario where an advanced AI could overwrite global musical history with a singular aesthetic. It estimates that converting 100 million commercial songs could be achieved by a large botnet in under two hours for approximately $26,667 in electricity, revealing the technical plausibility and severe cultural preservation risks posed by such AI.
Key insights
- Large-scale AI-driven musical transformation is technically and economically plausible, even with current open-source tools.
- Applying AI voice models to diverse musical traditions raises significant ethical concerns regarding cultural appropriation and homogenization.
- Digital musical history is highly vulnerable to rapid, large-scale manipulation, potentially leading to a "memory dictatorship."
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.