Technology news and discussion on TalkFuse.
Patreon stops asking AI bots not to scrape — and starts blocking them
Patreon has moved from politely asking AI companies not to scrape its content to actively blocking AI training bots, working with Cloudflare to enforce the crackdown. The shift matters because it marks a hardening of stance among content platforms towards generative AI firms, reflecting growing frustration that voluntary opt-out signals are being ignored while newer discovery features risk exposing more creator content to crawlers.
Previously, Patreon relied on robots.txt files, a standard but non-binding method of telling bots not to scrape a site, which some AI crawlers simply disregarded. Using Cloudflare's AI Crawl Control technology, Patreon says weekly access attempts by individual AI training crawlers dropped from thousands to zero during testing, though bots that index content to direct users back to Patreon will still be permitted. Product chief Drew Rowny said creators "deserve a meaningful say" in how their work is used by AI companies, positioning the move as giving creators control rather than forcing them to accept AI training as the price of reaching an audience.
In short
- Patreon now actively blocks AI training bots instead of just requesting opt-out
- Uses Cloudflare's AI Crawl Control; scraping attempts reportedly dropped to zero
- Indexing bots that redirect users to Patreon are still allowed
Steelman: both sides, in good faith
The strongest fair case each way — we don't pick a winner.
The case for
Content creators produce original work and deserve meaningful control over whether and how it is used for AI training. Voluntary opt-out systems have proven ineffective as AI companies ignore robots.txt signals, making technical enforcement necessary. Creators should not face a choice between sharing work online and surrendering control of it to AI firms; they deserve to retain agency over their intellectual property and how it generates value.
The case against
AI systems require vast amounts of training data to improve, and restricting access to publicly available internet content impedes beneficial technological progress. The distinction between search engine indexing and AI training both fundamentally involve processing publicly posted material, making selective blocking philosophically inconsistent. Overly restrictive policies raise development costs and disadvantage smaller companies, ultimately slowing innovation that benefits society broadly.
Explained simply
Patreon is a platform where creators—artists, writers, podcasters and others—post their work behind paywalls, with subscribers paying for access. In recent years, AI companies have been automatically scraping content from across the internet to train their systems, including from Patreon. Many creators have not consented to this and receive nothing in return, which has raised questions about whether these companies should be permitted to use their work this way.
Previously, Patreon and other platforms relied on a non-binding system: they posted requests asking web crawlers—automated tools that copy content—not to scrape their sites. AI companies often ignored these requests entirely. Many creators felt they had little choice but to accept their work being used to train AI systems without their knowledge or permission.
The conflict between AI companies' need for training data and creators' demand for control over their own work has become increasingly heated. Platforms are now pursuing stronger technical approaches to prevent scraping, reflecting frustration that polite requests do not work. How this disagreement is resolved may influence the future relationship between AI developers and creative professionals worldwide.
AI Art Celebrity Culture Entertainment Technology