Patreon has stopped politely requesting that AI crawlers stay away from creator content. On Thursday, the membership platform announced it’s working with Cloudflare to actively block AI training bots at the infrastructure level, according to TechCrunch AI. The shift is small in technical terms and large in what it signals.
Here’s the change in one line: robots.txt is a request, and Patreon decided requests weren’t working.
What Actually Changed
Since 2023, Patreon relied on robots.txt, the standard text file that tells bots which parts of a site they can touch. It’s honor-system infrastructure. Well-behaved crawlers read it and comply. Badly behaved ones read it and keep going.
Patreon is now using Cloudflare’s AI Crawl Control to enforce the policy instead of publishing it. The results from testing tell you everything about the old system:
- Individual AI training crawlers went from “thousands of attempts” per week to zero
- Those attempts were happening despite robots.txt telling them not to
- Blocking is selective, not blanket
That middle point deserves a beat. Thousands of weekly access attempts from crawlers that had already been asked to stop means the honor system wasn’t degraded. It was ignored.
“Consent shouldn’t depend on whether a scraper chooses to behave,” Patreon wrote in its announcement, as detailed in TechCrunch AI. That’s the cleanest one-sentence summary of why the entire publishing web is rethinking its bot policy right now.
Why Patreon, Why Now
Patreon had natural protection for years. Most creator content sits behind a paywall, and paywalls are decent crawler defense by accident.
Then Patreon built discovery features. The redesigned Home Feed and Quips, its short tweet-style posts, put more content in public view where crawlers can reach it. Growth features and scraping exposure move together. You can’t have open discovery and closed content at the same time without enforcement doing the work.
So the block isn’t a philosophical stance that arrived out of nowhere. It’s what happens when a company opens the front door and realizes who walks through it.
The Cloudflare Layer Is the Real Story
What stands out here is that Patreon didn’t build this. It bought it.
Cloudflare has been assembling an anti-scraping toolkit for months. It offers AI bot restrictions, and it launched Pay Per Crawl, a marketplace where sites can charge AI companies for access rather than block them outright. Earlier this month it went further, blocking “mixed-use” crawlers by default on ad-hosting pages. Mixed-use means a bot that both indexes for search and trains models. Cloudflare decided that ambiguity defaults to no.
That matters because Cloudflare sits in front of a large share of the web. When it changes a default, millions of sites change with it. Individual publishers negotiating with AI labs one at a time was never going to scale. Infrastructure-level enforcement does.
The Distinction Publishers Are Learning to Make
Patreon isn’t blocking every bot. It still allows crawlers that index pages and send users back to Patreon.
That’s the line the whole industry is converging on:
- Indexing bots bring traffic back. Traffic is the payment.
- Training bots absorb the content and return nothing.
The old web ran on an implicit deal where you gave search engines your content and got visitors. AI training broke that deal by taking the first half and skipping the second. Publishers aren’t anti-bot. They’re anti-free.
“On most of the Internet, creators have to accept AI training on their work just to reach and grow an audience,” said Patreon product chief Drew Rowny in the announcement. “Patreon has a different vision.”
What to Expect Next
If you run a site with content worth protecting, three things follow.
- Check what your robots.txt is actually accomplishing. Cloudflare’s crawler analytics will show you the gap between what you asked for and what happened.
- Decide where your indexing/training line sits before someone crosses it. Blanket blocking costs you search traffic.
- Watch Pay Per Crawl. Licensing beats blocking if the price is right, and pricing data is starting to exist.
The broader move is from policy to enforcement. Stating terms was never the same as having them respected, and a lot of publishers are finding that out at once. Expect more platforms to follow Patreon inside the next few quarters, especially ones with creator content and a paid tier to protect.
More details on the rollout are available at the original source.