Social media platform X has officially updated its developer agreement to strictly prohibit third parties from utilizing the platform’s data to train large language models (LLMs). This strategic move, implemented this Wednesday, effectively closes the door on unauthorized AI scraping of the site’s vast user-generated content.
New Restrictions on API Usage
The company introduced a specific clause under the “Reverse Engineering and other Restrictions” section of its developer policy. The updated text is explicit: “You shall not and you shall not attempt to (or allow others to) use the X API or X Content to fine-tune or train a foundation or frontier model.”
Consolidating Data for xAI
This policy pivot follows the acquisition of X by Elon Musk’s artificial intelligence firm, xAI, in March. Industry analysts suggest the decision aims to protect the platform’s proprietary data, ensuring that competitors cannot leverage X’s real-time information to build or improve their own competing AI systems without a formal licensing agreement.
A History of Shifting Data Policies
The platform’s stance on AI training has been fluid over the past year. In 2023, X modified its privacy policy to explicitly allow the use of public data for its own AI training purposes. By October, the company had initially opened the gates to allow third parties to train their models using X data. This week’s update marks a significant reversal of that previous openness.
Industry-Wide Trend Toward Data Protection
X is not alone in its efforts to restrict AI access to proprietary content. Other major platforms are increasingly erecting walls to protect their ecosystems. Reddit has implemented similar safeguards to block AI crawlers, and companies like The Browser Company have recently introduced restrictive clauses in their terms of use for AI-focused tools like the Dia browser to prevent unauthorized training usage.