Patreon hardens its bot policy: from robots.txt to effective blocking

TechCrunch documents on July 17 a shift at Patreon - the platform stops politely asking AI scrapers to leave, it blocks them.

The fact

TechCrunch publishes on July 17, 2026, "Patreon stops asking AI bots not to scrape - and starts blocking them." The move is clear: the platform shifts from a declarative policy (robots.txt, terms of service) to an enforcement policy (effective blocking at the infrastructure level). This is a technical and legal shift.

Our analysis

The issue goes beyond Patreon. Robots.txt is a 30-year-old social contract that has never had legal force—the LLM models have made it obsolete in practice. The shift to effective blocking does two things: it enforces the rights holder's will without going to court, and it makes the violation technically traceable when it occurs. Expect content creators (New York Times, Reddit, Stack Overflow) to harden their own implementations. The regulatory framework (AI Act on copyrights) is accelerating the movement.

To watch

Reactions from OpenAI, Google, Anthropic; first judicial decisions citing Patreon as a precedent for clear expression of refusal.