Elon Musk's X giving users a chance to opt out of Grok AI training, but it's not the default

Elon Musk's X (formerly Twitter) is giving users more control over how their data is used to train its AI chatbot, Grok. However, the default setting remains in favour of data collection, requiring users to actively opt out if they wish to prevent their posts and interactions from being used for AI training.

"All X users have the ability to control whether their public posts can be used to train Grok, the AI search assistant," the company announced in an X post on Friday. "This setting is available on the web platform and will soon be rolled out on mobile."

Users can also prevent their data from being used by making their X accounts private.

Grok, an AI chatbot currently available to premium X subscribers, was launched in November 2023 as a competitor to OpenAI's ChatGPT. The chatbot, developed by Musk's AI startup xAI, has been trained extensively on data scraped from X.

Musk has been open about his intentions to utilise X data for Grok's development. The chatbot already leverages real-time information from X posts to summarise news events and answer user queries.

This move to allow users to opt out of Grok training comes amid growing concerns about data privacy and the use of personal information for AI development. However, the fact that opting out requires a manual change in settings, rather than being the default, highlights the tension between user privacy and the vast amounts of data required for training powerful AI models.