On October 17, 2024, X (previously known as Twitter) revised its Privacy Policy to allow third-party corporations to use X user postings to train their artificial intelligence (AI) systems. This move implies that, unless users explicitly opt out, third-party collaborators will be able to access and use platform data to train AI models.
This change follows the activities of X’s owner, Elon Musk, who has previously educated xAI’s Grok AI chatbot on X data. This decision prompted an inquiry by the European Union’s top privacy authority. However, until today, X has not changed its privacy regulations to reflect that third parties, other than xAI, might utilize the platform’s data for AI training.
The amended policy appears to be an attempt by X to make cash by selling or licensing data to AI businesses, a tactic similar to that pursued by other platforms such as Reddit and media organizations.
Key Changes in the Privacy Policy
The amended Privacy Policy, notably Section 3, now includes a phrase concerning “Sharing Information.” It outlines how user data may be shared with third parties, who may use the data for their own objectives, such as training AI systems. Users can choose not to share their data, but the policy does not specify how. Currently, X’s settings page enables users to manage data-sharing with xAI’s Grok and some “business partners,” but it’s unclear whether and how customers may prohibit third-party AI businesses from utilizing their data until the policy goes into effect on November 15.
Another notable modification to the Privacy Policy is how long X retains user data. X previously said that it will preserve “personally identifiable” data for a maximum of 18 months. The platform now shows that different types of data will be retained for differing periods of time, based on variables such as legal constraints and safety concerns. For example, content uploaded by users will be retained for the duration of the user’s account or until the user deletes it.
X also stated that even if a user deletes their postings from the platform, copies of the data may still persist elsewhere. Search engines and other third parties may retain the data in accordance with their respective privacy rules. This remark might apply to circumstances where data has already been exploited by AI companies.
Monetizing Data and Scraping Fees
X is also introducing new rules for organizations that scrape or gather large amounts of data from the platform. A new section in X’s Terms of Service, called “Liquidated Damages,” sets penalties for organizations that access more than 1,000,000 posts in a 24-hour period. These organizations could be charged $15,000 for every 1,000,000 posts they access.
This new policy could help X monetize its data as the company faces challenges from advertisers leaving the platform and a subscription service that has not gained much traction.
While the amended policies raise privacy issues for users, X did not react to requests for feedback on the revisions.
This information is based on Sarah Perez’s TechCrunch report, which was released on October 17, 2024. You can check out the full article here.
I’m Voss Xolani, and I’m deeply passionate about exploring AI software and tools. From cutting-edge machine learning platforms to powerful automation systems, I’m always on the lookout for the latest innovations that push the boundaries of what AI can do. I love experimenting with new AI tools, discovering how they can improve efficiency and open up new possibilities. With a keen eye for software that’s shaping the future, I’m excited to share with you the tools that are transforming industries and everyday life.