Without notice, X is now scraping posts and interactions to train its third-rate AI product Grok

Heads up! Without warning or notice, Elon Musk’s social platform X (formerly Twitter) has quietly begun scraping users’ posts and interactions with Grok to train the generative AI chatbot used by dozens. The setting to allow data for training is enabled by default. Thankfully, there’s a way users can opt-out…hidden only in the web version of X.

9to5Mac Security Bite is exclusively brought to you by Mosyle, the only Apple Unified Platform. Making Apple devices work-ready and enterprise-safe is all we do. Our unique integrated approach to management and security combines state-of-the-art Apple-specific security solutions for fully automated Hardening & Compliance, Next Generation EDR, AI-powered Zero Trust, and exclusive Privilege Management with the most powerful and modern Apple MDM on the market. The result is a totally automated Apple Unified Platform currently trusted by over 45,000 organizations to make millions of Apple devices work-ready with no effort and at an affordable cost. Request your EXTENDED TRIAL today and understand why Mosyle is everything you need to work with Apple.

In a series of viral X posts on Friday, users began to notice that the platform added a new “Data sharing” toggle within settings that explains any posts made on the platform, as well as interactions, inputs, and results with Grok are now being used for “training and fine-tuning.”

Up until now, Grok has only been trained on a hodgepodge of publicly available text from the Internet, with the capabilities and data sets nowhere near that of ChatGPT or Co-Pilot. X claims that Grok’s access to real-time public X posts allows the chatbot to respond to user queries with up-to-date information and insights on a wider range of topics.

About Security Bite: Security Bite is a security-focused column on 9to5Mac. Every week, Arin Waichulis sheds light on the latest in data privacy, vulnerabilities, or emerging threats within Apple’s vast ecosystem of over 2 billion active devices.

This also means that when you interact with Grok, your interactions, inputs and results may be used to train and enhance the system’s performance. This allows us to:

• Enhance Grok’s understanding of human language and communication.
• Improve Grok’s ability to provide accurate, relevant, and engaging responses.
• Develop Grok’s sense of humor and wit to make interactions more enjoyable.
• Ensure Grok remains politically unbiased and provides balanced answers.

Skeptical? So am I.

I don’t want to get too in the weeds here, but the X today isn’t the Twitter that existed when Elon Musk acquired the platform in June 2022 for ~$44B. Under the veil of freedom of speech, the social platform’s moderation has fallen by the wayside. Extreme ideologies, hate, spam, and factually incorrect information run rampant. The quality of training data is just as important as the data itself.

If Grok is trained on data containing hate speech or far-leaning beliefs, it could inadvertently replicate these biases leading to the AI generating harmful or misleading content. If there’s one platform I trust least to handle something like this it’s X.

How to opt-out

As of now, you can only opt out of Grok scraping your posts on the web version of X. The option does not exist in the mobile app. If you’re unable to access the opt-out quick link here, follow the steps below:

• On X web select “Privacy & Safety”
• Under “Data sharing and personalization,” select “Grok”
• Toggle off the option “Allow your posts as well as your interactions, inputs, and results with Grok to be used for training and fine-tuning.”

At the end of the day, there’s a not-so-old adage: If you’re not paying for the product, then you are the product.