Mastodon Bans AI Scraping, Updates Terms to Block Model Training & Raise Age Limit

If you’re someone who thinks your data should be off-limits to AI training bots, Mastodon has some news you’ll probably appreciate. Mastodon has updated its terms of service to ban AI scraping.

“We explicitly prohibit the scraping of user data for unauthorized purposes, e.g. archival or large language model (LLM) training,” the platform noted. “We want to make it clear that training LLMs on the data of Mastodon users on our instances is not permitted.”

The updated terms go into effect July 1 and aim to block any kind of automated data extraction. That includes bots, crawlers, scrapers, offline readers—basically anything that pulls data in bulk without a human behind it. If it’s not for standard search or browser functions, it’s off-limits.

There is one catch, though. These rules only apply to Mastodon.social, which is just one server in the larger fediverse. Since Mastodon is decentralized, other instances can still set their own rules. So technically, AI companies could still scrape from other servers—unless those servers also ban AI scraping.

The move comes not long after X (formerly Twitter) updated its own terms to ban model training, following a growing trend. Reddit, OpenAI, and even The Browser Company have also laid down similar rules to prevent outside companies from turning public content into training fodder.

Mastodon’s update also introduces a new age restriction. The platform is raising its minimum age from 13 to 16, and this one applies globally.