Social networks are tightening their terms of service to combat scrapers and bots that crawl the site to train AI models. A few days after Elon Musk’s Company X updated its terms of service to explicitly ban AI model training, decentralized social network Mastodon today updated its rules to also ban any model training.
“We explicitly prohibit the extraction of user data for unauthorized purposes, such as archiving or training large language models (LLMs). We want to make it clear that training LLMs on Mastodon user data on our instances is not permitted,” the company said in an email sent to Mastodon users.
The new terms, which will be applied to the social network from July 1, contain legal language prohibiting any data extraction and development of an automated system.
“Use, run, develop or distribute any automated system, including, without limitation, any spider, robot, spinner, scraper, offline reader or any data mining or similar data collection tools to access the Site, except as may be the result of a standard search engine or internet browser and local caching or for human viewing and interaction with the Content on the Site,” the terms state.
It is important to note that these terms apply only to the Mastodon.social server, which is only one of the instances in the fediverse distributed network. This means that scrapers can still pull data from other servers and use it to train AI models, unless they explicitly prohibit it in their terms of service.
Other platforms, including OpenAI, Reddit, and The Browser Company, have added similar clauses to their terms of service to prevent other companies from training models.
In addition to this change, Mastodon is also introducing a new age limit for users – 16 years old. Previously, the social network had an age limit of 13 for users in the US, but now it is changing the age limit worldwide.
