X now bars training of AI models using its content
What's the story
Social media platform X has updated its developer agreement to prevent third parties from using the platform's content for training large language models (LLMs). The new restriction was added under "Reverse Engineering and other Restrictions," a subsection of use restrictions. The clause states that users cannot use the X API or content to fine-tune or train a foundation or frontier model.
Acquisition impact
Policy change follows xAI's acquisition of X
The policy change comes after Elon Musk's AI company, xAI, acquired X in March. This move is not surprising as xAI would not want its competitors to have free access to data from the social platform without a sales agreement. The updated policy also aligns with other tech companies like Reddit and The Browser Company that have implemented similar measures against AI crawlers.
Future prospects
Potential AI training deals with 3rd-party companies
The updated policy from X could open doors for potential AI training deals with third-party companies. This is similar to a deal Reddit made with Google. Despite the new restriction in its developer agreement, X's privacy policy still allows third-party "collaborators" to train AI models on the site's data unless users opt out.