On Wednesday, social network X (formerly Twitter) updated its privacy policy to say it will now allow third-party “collaborators” to train AI models on X data unless users opt out. X owner Elon Musk trained xAI's Grok AI chatbot on X user data, leading to an investigation by the EU's chief privacy regulator, but the company said the data could also be used by third parties. had not yet amended the policy to indicate that this could be the case.
This addition to the policy suggests that X, like Reddit and various media organizations, is considering licensing data to AI companies as a potential new revenue stream.
In Section 3 of the updated Privacy Policy, entitled “Information Sharing,” X has added a paragraph detailing how X uses your data and how you can opt out.
It says:
“Third Party Collaborators. Depending on your settings or if you decide to share your data, we may share or disclose your information with third parties. If you do not opt-out, In some cases, the recipient of the information may use the information for its own independent purposes other than those described in X's Privacy Policy, whether generative or otherwise. Objectives include training intelligent models.
This policy points to the X settings page, but doesn't specify where users can go within settings to toggle data sharing off. Currently, in the “Privacy & Safety” section of Settings, users can turn on or off xAI's data sharing with Grok and other “business partners,” the latter of which allows X to “operate and improve our business.” “Companies with whom we may potentially cooperate for the purpose of achieving this goal.'' Not other AI providers.
This may be because our updated Privacy Policy does not take effect until November 15th, at which time an opt-out option may be added. (We hope.)
The company also says it will retain your “profile information and content for the life of your account” and other “personally identifiable data collected when you use our products and services for up to 18 months.” I deleted the phrase mentioned. ”
Instead, the new section states that “We retain different types of information for different lengths of time depending on the circumstances.” And for security reasons. As an example, it states that usage information, such as “the content you post” and your interactions with the content of others, is retained “for the life of your account or until such content is deleted.”
We've also added a note to this policy to notify users that public content may exist elsewhere even after it has been removed from X. This could cover data ingestion by AI providers, and X adds that “search engines and other third parties could retain copies of posts for longer.” , will be created under its own privacy policy even after deletion or expiration on X date. ”
Separately, X added a new “liability” section to its updated terms of service, stating that organizations that scrape content will be liable for damages. Specifically, “if you request, view, or access more than 1 million posts (including reply posts, video posts, image posts, and other posts) in a 24-hour period,” your organization will receive 1 million posts. Company X says it will be charged $15,000 per month. .
The move to monetize XData follows advertiser withdrawals, boycotts and a yet-to-be-ubiquitous subscription feature, leaving the company with a need for new ways to pay.
X did not respond to requests for comment.