X Prohibits Third Parties from Using Platform Content for AI Model Training
The updated policy aims to restrict the use of data from the social media platform for training large language models.
On June 6, 2025, the social media platform X implemented a significant update to its developer agreement, prohibiting third parties from utilizing the platform's data and content for training large language models.
The revision, outlined in the 'Reverse Engineering and Other Restrictions' section, explicitly states that developers 'MAY NOT, NOR ATTEMPT TO (NOR ALLOW OTHERS TO DO SO) ADJUST OR TRAIN A FOUNDATION OR FRONTIER MODEL.' This policy change follows two months after X was acquired by the artificial intelligence company XAI, owned by Elon Musk.
Musk had previously stated that the acquisition aimed to 'combine data, models, computing, distribution, and talent.' He emphasized that the merged company would provide 'smarter and more meaningful experiences to billions of people,' while remaining committed to its primary mission of seeking truth and promoting knowledge.
The announcement comes amid broader discussions on the implications of using social media content in training AI, as platforms worldwide grapple with the management of user data and privacy concerns.
The revision, outlined in the 'Reverse Engineering and Other Restrictions' section, explicitly states that developers 'MAY NOT, NOR ATTEMPT TO (NOR ALLOW OTHERS TO DO SO) ADJUST OR TRAIN A FOUNDATION OR FRONTIER MODEL.' This policy change follows two months after X was acquired by the artificial intelligence company XAI, owned by Elon Musk.
Musk had previously stated that the acquisition aimed to 'combine data, models, computing, distribution, and talent.' He emphasized that the merged company would provide 'smarter and more meaningful experiences to billions of people,' while remaining committed to its primary mission of seeking truth and promoting knowledge.
The announcement comes amid broader discussions on the implications of using social media content in training AI, as platforms worldwide grapple with the management of user data and privacy concerns.
