Elon Musk's X is using your tweets to train its AI, Grok. Here's how to opt out.

• Elon Musk's X activated a default setting to train Grok, its AI chatbot, on user data.
• Users can disable data sharing in the 'Privacy and Safety' tab on X.
• Grok, a ChatGPT competitor, is available via premium subscription.

Your unfiltered thoughts are about to become training data for one of the most unhinged AI chatbots on the market.

X has activated a setting that, by default, gives it access to users' posts, interactions, and inputs as training data for Grok, the company's ChatGPT competitor.

If you want to disable it, here's a step-by-step guide:

Once on X, head to the menu on the left side and click "More."

Then, select the "Settings and privacy" tab at the bottom.

Under that tab, navigate to the "Privacy and safety" option.

Then, scroll down and uncheck the box next to Grok.

Grok, which first became available to X users last year, is now available to all users via a premium subscription. It's been billed by X owner Elon Musk as loving sarcasm and having a "little humor" in its responses — partly because it has real-time access to a trove of user posts.

The bot comes in two modes: a "regular mode" that provides straightforward answers and a "fun mode" that amps up the sarcasm and humor.

Last year, Musk said in Spaces that he wants Grok to be known as the funniest AI. "So I mean if you ask Grok to provide a vulgar roast, it's really good," he said.

It's yet to be seen just how funny Grok will be if everyone opts out of allowing it to train on their data.