Patreon stopped asking crawlers nicely and started blocking them

TechCrunch reports that Patreon has started actively blocking AI bots that scrape creators’ content for model training without permission through Cloudflare. The company is building on measures it introduced in 2023, but moving from robots.txt instructions to technical enforcement through AI Crawl Control.

Robots.txt no longer works as a fence around creators

Patreon says AI scraping has become more sophisticated since its first deterrence measures. Much creator content has long sat behind a paywall, but newer discovery features, including the redesigned Home Feed and short posts called Quips, can expose more material to crawlers.

TechCrunch cites a concrete test from Patreon: weekly access attempts by individual AI training crawlers reportedly went from thousands to zero. That is a stronger signal than another public statement about consent.

Creator platforms need enforcement, not just etiquette

For platforms built on creator content, this is a practical shift in power. Robots.txt works only when a crawler chooses to respect the rules. Active blocking at the infrastructure layer moves the decision from a request to an access control.

Cloudflare fits into this with a broader publisher toolkit, including controls for AI bots and a Pay Per Crawl model. Patreon also says it will still allow bots that index pages and send users back to the platform.

Bot blocking will not settle the training data market

A technical barrier helps, but it does not fully answer who gets to monetize creators’ work. Crawlers can change identity, content can leak through users and some value will move into contracts between platforms and AI companies.

There is also a dependency risk. If access rules for the web move toward a few large infrastructure providers, creators gain protection but also another intermediary between themselves and their audience.

Licensing terms and exceptions are the real test

The next signal is how Patreon handles exceptions: which bots are allowed, under what terms and whether creators get their own controls. That will show whether this is author protection or a stronger negotiating position for the platform.

It is also worth watching whether other platforms with closed or semi closed content follow. If they do, the era of voluntary non scraping on the web will end faster than AI labs would like.

Lilith's verdict

Robots.txt was a sign on the door. Patreon is now installing a turnstile and asking who has a badge. For creators, that is late, but still better than leaving the warehouse open overnight.