X gave Grok AI automatic permission to train using your data—Here’s how to stop it

 

Twitter users have been automatically opted in to sharing their data with the tool.

BY Sarah Bregel

In more questionable X (formerly Twitter) news, the social media site recently activated a new setting that will allow AI to use your data to train. Basically, X opted you into handing out your posts to its Grok AI tool, a rival of ChatGPT, without your permission. 

That means that even if you aren’t using AI, the robots can pull from anything you’ve ever posted on the site. While Grok is a premium X feature, that doesn’t mean you have to have a paid account for Grok to access your account. The new hidden setting will allow your posts to train the tools, even if you aren’t using Grok.

Luckily, you can easily disallow the robots from snagging your content and spitting it back out in just a few clicks.

Since X gave its AI tools automatic permission to use your posts, you’ll have to update your settings to tell it not to. To make it slightly trickier, you can’t do it on your phone—only on the web.

Here’s how:

Open up the settings page on X. You can find it here. Then, go into “Privacy and safety.” Select “Grok,” then uncheck the Data sharing box that says “Allow your posts as well as your interactions, inputs, and results with Grok to be used for training and fine-tuning.” That’s it! 

It feels a little, well, icky to see that box has been checked for you. However, it only takes a second to turn it off, and it feels moderately empowering to do so. Unless, of course, you don’t mind the bots using your posts. Then, by all means, leave that box checked.

No matter how you feel about AI and its increasingly active role in our lives, there have been significant concerns surrounding X’s AI tool. For one, it has made up multiple news stories, which X has promoted on the trending news page Explore. One was about the total solar eclipse. Another headline incorrectly claimed that Iran had launched a retaliatory attack on Israel. 

Elon Musk has expressed hope about Grok being the “anti-woke” AI. However, with Grok using X to train, there are fairly real concerns about how much blatantly false information it will gather and spit back out as “fact.”

As long as you adjust your settings, at least you won’t be part of the problem. How reliable the Grok’s of the world are, though, is to be determined.

 

ABOUT THE AUTHOR

Sarah Bregel is a writer, editor, and single mom living in Baltimore, Maryland. She’s contributed to NYMag, The Washington Post, Vice, In Style, Slate, Parents, and others. 


Fast Company

(2)