The Register: Training AI on Mastodon posts? The idea’s extinct after terms updated

Jun 18, 2025

—

Source URL: https://www.theregister.com/2025/06/18/mastodon_says_no_to_ai/
Source: The Register
Title: Training AI on Mastodon posts? The idea’s extinct after terms updated

Feedly Summary: Such rules could be tricky to enforce in the Fediverse, though
Mastodon is the latest platform to push back against AI training, updating its terms and conditions to ban the use of user content for large language models (LLMs).…

AI Summary and Description: Yes

Summary: The text highlights Mastodon’s recent policy update that prohibits the use of user content for training large language models (LLMs). This reflects an ongoing tension within the Fediverse regarding AI training practices and user privacy.

Detailed Description: The content discusses Mastodon, a decentralized social media platform, taking a stand against the usage of its users’ data for AI training purposes. This development is notable for several reasons:

– **Policy Update**: Mastodon has updated its terms and conditions specifically to prevent the use of user-generated content for training LLMs.
– **User Privacy**: The move signals an active concern for user privacy and data ownership within platforms that contribute to the Fediverse, a network of interconnected social media platforms.
– **AI Training Debate**: The text reflects a growing debate around ethical AI practices, particularly regarding transparency and consent in data usage.
– **Impact on AI Development**: By restricting the use of user-generated content, Mastodon and similar platforms could potentially limit the training data available for LLMs, impacting the development of AI technologies.

Overall, this text is significant for professionals in AI security, infrastructure security, and privacy, as it addresses the implications of policies aimed at protecting user data in the context of evolving AI practices. It underscores the need for compliance with privacy concerns and regulatory frameworks while developers and organizations leverage user data for AI advancements.

1 2 2025 5 a Act addresses advancement advancements after AI AI advancements AI development AI security AI technologies and art as Bi by C CERN CI CIA co compliance concerns Condi Consent content Context core D data data ownership data usage de decentralized developer developers development e Entra ethical ethical AI ethical AI practices event Fediverse for framework frameworks g Gen generated Generated Content GIS Go H high Highlight http HTTPS implications in infrastructure infrastructure security inter io k l language language model language models large large language model large language models Large Language Models (LLMs) led Li llm llms lm M Mastodon media Mila Mode model models N network NGO no o of on OPM organization organizations ory oS over ownership platform platforms policies policy policy update post potential practices pre privacy privacy concerns pro professionals ps R rag rate RCE regulatory regulatory framework regulatory frameworks Ro RoT row s sec security Sig Signal Sim SoC social social media social media platforms source specific SSE T taking tech technologies ted test text the to Tor TP training training data training large language models training practices transparency two under up update US usage use user user content user data user privacy user-generated content Users V Wi x