Patreon moves to block AI scrapers instead of asking for compliance

The company’s previous reliance on robots.txt files proved inadequate as AI scrapers routinely bypassed these standard instructions. Patreon reported that before implementing its new defenses, individual bots made thousands of unauthorized access attempts per week. The new infrastructure-level block reduces these attempts to zero, ensuring that consent is no longer optional for developers harvesting data for model training.

This aggressive stance follows the introduction of public-facing features like the redesigned Home Feed and Quips, which increased the platform’s exposure to crawlers. While the company is actively barring training-focused bots, it will continue to permit crawlers that index pages specifically to drive traffic back to creators. Drew Rowny, Patreon’s product chief, emphasized that this policy represents a push for creator agency, contrasting the platform's approach with the wider internet where scraping is often a prerequisite for audience growth.