Social network X (formerly Twitter) recently activated a setting that gives it permission to train Grok AI on user tweets. All X users are opted in by default, with X failing to notify customers about the change.
The hidden setting gives X permission to use all posts, interactions, inputs, and results for "training and fine-turning" Elon Musk's Grok AI model.
To continuously improve your experience, we may utilize your X posts as well as your user interactions, inputs and results with Grok for training and fine-tuning purposes. This also means that your interactions, inputs, and results may also be shared with our service provider xAI for these purposes. Learn more
While X turned the setting on for all users by default, it can be turned off on the website. Here's how:
- Open up X using a web browser on a Mac or PC and log in.
- Click on More.
- Choose Settings and privacy.
- Click on Privacy and safety.
- Click on Grok.
- Uncheck the setting that allows Grok to use your content.
This privacy setting is not available through the X app, so it must be disabled using the desktop version of X, though an option to turn it off on mobile devices is supposedly coming soon.
All X users have the ability to control whether their public posts can be used to train Grok, the AI search assistant. This option is in addition to your existing controls over whether your interactions, inputs, and results related to Grok can be utilized. This setting is… — Safety (@Safety) July 26, 2024
Turning on a data collection feature surreptitiously and automatically opting in users without their consent is underhanded, and many X users are not happy with the social network's decision.
Sometime in the last month, X updated its Help Center to change the wording around Grok. It previously said that "user interactions, inputs, and results" would be used for Grok training, but now the wording also includes X posts.
To continually improve the AI system and provide a better user experience, we may utilize user interactions, inputs and results for training and fine-tuning purposes. This means that when you interact with Grok, your interactions, inputs and results may be used to train and enhance the system's performance.
Today's wording:
To continuously improve your experience, we may utilize your X posts as well as your user interactions, inputs and results with Grok for training and fine-tuning purposes. 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.
Grok has previously been able to access "real-time public X posts" to respond to user queries, but it is not clear if X posts were being used for training. The opt-out toggle seems to have been enabled around May, so it's possible that's when user posts were first used for improving the chatbot.
X says that user interactions with the chatbot and X posts are used for the following:
- 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.
Elon Musk claimed earlier this week that Grok was being trained on "the most powerful AI training cluster in the world." Grok is currently available to X premium subscribers.
