Social networks strengthen their terms of service in relation to scalders and bots that crawl on the website for training AI models. A few days after the X Musk X updated its conditions to clearly prohibit the training of the AI model, the decentralized social network of Mastodon has updated its own rules today to prohibit all kinds of model training.
“We are clearly prohibiting the scraping of user data to unauthorized purposes, e.g. archival training or a large language (LLM). We want to explain that LLM training on the data of Mastodon users in our cases is not allowed,” said Mastodon We -mail sent to users.
. new conditionswhich will apply to the social network from July 1, they have a legal language that prohibits data extraction and developing an automatic system.
“Using, starting, developing or distributing any automated system, including without restrictions, any spider, robot, cheating tools, scraper, offline reader or any data exploration or similar data collection tools to gain access to instance, except for each case, with the exception of any case, which may be the result of a standard search engine or a local armor local human fragrance or human interaction and interaction with content, “.
It should be noted that these conditions apply only to Mastodon.Social Server, which is only one of the instances in the Fediverse distributed network. This means that scrapers can continue to separate data from other servers and use them to train AI models if they do not clearly block it in their terms.
Other platforms, including OpenaiReddit and The Browser Company, They added similar clauses to their rules to prevent other model training companies.
In addition to this change, Mastodon also enforces a new age of 16 for users. The social network had a limit of 13 for users in the US, but changes the age limit around the world.
