Patreon moves from asking AI bots to stay away to actively blocking them

Lead

Patreon is taking a harder line against AI scraping. According to TechCrunch, the creator membership platform is expanding its work with Cloudflare to directly block bots that try to access creators’ work for AI model training without permission.

The change is notable because Patreon is moving beyond the traditional web practice of using robots.txt files to tell crawlers what they may or may not do. Robots.txt can express a site’s preference, but it depends on crawlers choosing to comply. Patreon’s new approach is to enforce those preferences at the access-control level.

Key points

• From request to enforcement: Patreon had already taken steps in 2023 to deter AI crawlers. The company now says scraping has become more sophisticated, and simply asking bots not to collect content is no longer enough.
• Cloudflare’s role: Patreon will use Cloudflare’s AI Crawl Control technology to update its AI policies and enforcement tools. Cloudflare has been building products for publishers to restrict AI bots, including tools such as Pay Per Crawl.
• Why the timing matters: Much of Patreon’s creator content has long been protected by paywalls. But newer discovery surfaces, including a redesigned Home Feed and Quips, could make more material visible to crawlers.
• Not all bots are treated the same: Patreon says it will continue to allow bots that index pages, organize information, and direct users back to Patreon. The target is unauthorized scraping for AI training.

Why it matters

The move reflects a growing tension across the web: creators want reach and audience growth, but that does not mean they automatically consent to having their work used to train AI models. For years, robots.txt served as a lightweight social contract between websites and crawlers. In the AI era, many platforms are concluding that a social contract is not enough if some scrapers ignore it.

For creators, Patreon’s approach makes consent more than a policy statement. It turns the platform’s rules into a technical barrier. For AI companies, it signals that access to high-quality creator content may become more constrained, more negotiated, and potentially more costly.

The broader shift is toward a more segmented web, where search indexing, user referral, and AI training are treated as separate uses rather than one broad category of crawling. Patreon’s decision will not settle the larger debate over AI training data, but it does mark an important step: platforms are moving from saying “please don’t scrape” to deciding, technically, who gets in.

Source: TechCrunch AI