Academic Study Reveals Generative AI Flooding Self-Publishing Market, Diluting Revenue Despite Growing Sales Share
Key Takeaways
- ▸AI-generated books comprise a large share of self-published catalogs but capture disproportionately smaller sales volume compared to human-authored works
- ▸Despite quality concerns, AI books are growing their market share over time and increasingly winning scarce top-rank positions previously held by non-AI books
- ▸The self-publishing market is adding books 19.2x faster than it is adding revenue, indicating market saturation and reader preference for human-authored content
Summary
A comprehensive arXiv research paper analyzing 14,419 self-published genre-fiction books sold on Amazon from 2023 to 2026 reveals that generative AI-produced books are reshaping the market through sheer volume rather than quality. The study, which used full-text AI detection and matched books to daily sales records, found that while AI-generated books (containing >25% detected AI text) comprise a substantial portion of the self-published catalog, they still capture a smaller share of actual sales compared to human-authored works.
Despite their lower per-book quality reputation, AI-generated books are steadily gaining market share over time and increasingly capturing top-rank positions previously dominated by non-AI books. The data shows a striking divergence: the number of books with observed sales in a quarter grew 19.2-fold between 2023 and 2026, while quarterly revenue grew only 8.9-fold. This means the market added selling books at more than double the rate it added revenue, with revenue per selling book declining across most genres.
The research reveals that human-authored books are losing ground most severely in genres with high AI diffusion, particularly where Kindle Unlimited availability is high. The paper's findings have direct implications for copyright law, as top-selling AI-generated books draw on more distinctive language from existing published works than human-authored books do, with this correlation rising alongside revenue—a pattern not observed in traditionally-written works.
- The rise of AI-generated content in publishing raises critical fair-use and copyright questions, with implications for how existing laws protect authors and creative workers
Editorial Opinion
This rigorous quantitative analysis provides compelling evidence that generative AI is reshaping creative markets through scale and saturation rather than by producing higher-quality work. The data suggests that while AI-generated books are individually dismissed as 'slop,' their collective volume is eroding the viability of human authors in self-publishing—a concerning trend that demands urgent attention from copyright law, publisher practices, and platform policies that govern content distribution and discovery.
