UPDATE 11/15: Thinking about dropping X? A change implemented today may help push you toward the exit. As we explained in October (see below), the company updated its terms of service today to officially say it uses public tweets to train its AI. The new doc reads:
“By submitting, posting or displaying Content on or through the Services, you grant us a worldwide, non-exclusive, royalty-free license (with the right to sublicense) [and] you agree that this license includes the right for us to (i) analyze text and other information you provide and to otherwise provide, promote, and improve the Services, including…for use with and training of our machine learning and artificial intelligence models, whether generative or another type.”
As we outline in this explainer, you can opt out of this training on the desktop. As of 10 a.m. ET on Friday, the option was still available. On X.com, navigate to Settings and privacy > Privacy and safety > Grok and uncheck the box that says Allow your posts as well as your interactions, inputs, and results with Grok to be used for training and fine-tuning.
If that’s not enough, here’s how to delete your tweets and/or your X account.
Original Story 10/18:
An upcoming revision to X’s terms of service will make a recent plot twist at Elon Musk’s social platform a black-and-white contractual reality: Your posts can officially serve as food for X’s Grok chatbot and any other AI systems that X might create in the future.
This practice has been documented at X since the company quietly updated a help file in July to note that your public posts (as well as your interactions with Grok itself) can be used to train X’s AI models. That news, in turn, came a year after Musk revealed that his AI project would be trained on people’s tweets.
The new terms of service, which go into effect on Nov. 15, spell out the new policies. Under section 3, “Content on the Services,” the terms require users to agree that X has the right to “analyze text and other information you provide and to otherwise provide, promote, and improve the Services, including, for example, for use with and training of our machine learning and artificial intelligence models, whether generative or another type.”
It’s unclear whether that language means X will remove the current opt-out option that allows users to disable Grok’s AI scraping. X did not immediately respond to a query emailed to its press-relations address, and a look over Musk’s 100 most recent posts and replies did not reveal further details.
That’s not the only change to X’s ToS that’s raising questions. Under section 5, a new “Liquidated Damages” section gives the platform the right to fine users who use it too much.
“Protecting our users’ data and our system resources is important to us,” the section says before declaring that “you will be jointly and severally liable to us for liquidated damages as follows for requesting, viewing, or accessing more than 1,000,000 posts (including reply posts, video posts, image posts, and any other posts) in any 24-hour period,” at a rate of $15,000 for each million posts.
If you keep doing that, this section adds, X reserves the right to enact “injunctive and/or other equitable relief, in addition to monetary damages.”
This policy is presumably meant to dissuade competitors from scraping X to train their own AI models, but some warn that it could also block researchers from using automated scraping tools for academic purposes.
The Knight First Amendment Institute at Columbia University called this “a disturbing move” in a statement posted Thursday.
“The public relies on journalists and researchers to understand whether and how the platforms are shaping public discourse, affecting our elections, and warping our relationships,” Knight litigation director Alex Abdo said. “One effect of X Corp.’s new terms of service will be to stifle that research when we need it most.”
Recommended by Our Editors
The revised terms of service also include new language that reflects X’s recent move of its headquarters from San Francisco to the Austin suburb of Bastro. The company will soon require that all legal disputes be brought in “the U.S. District Court for the Northern District of Texas or state courts located in Tarrant County.”
Why? That federal court is based in Dallas (Austin lies within the Western District of Texas) and the county it resides in is next door to Dallas. But Musk and X have already gone out of their way to file lawsuits there against a now-defunct group of advertisers and the liberal advocacy group Media Matters for America—both times choosing a small division of the Northern District in which a single particularly conservative judge, Reed O’Connor, hears cases.
Lastly, the company also plans to change how blocking works on X. A post from the XEng account on Oct. 17 states that the block function will soon no longer hide your posts: “If your posts are set to public, accounts you have blocked will be able to view them, but they will not be able to engage (like, reply, repost, etc.).” A second post in that thread suggested that this move was a response to people using the block feature “to share and hide harmful or private information about those they’ve blocked.”
As of this publication, using the block option still yielded a dialog saying that blocked users “will not be able to follow you or view your posts.” But with Musk having publicly said the old block function “makes no sense,” it’s likely only a matter of time before the change takes effect.
Editors’ note: We updated this post with added commentary about the litigated-damages clause.
Get Our Best Stories!
Sign up for What’s New Now to get our top stories delivered to your inbox every morning.
This newsletter may contain advertising, deals, or affiliate links. Subscribing to a newsletter indicates your consent to our Terms of Use and Privacy Policy. You may unsubscribe from the newsletters at any time.
About Rob Pegoraro
Contributor
