Elon Musk Used Twitter Users as Test Subjects! This Is How He Trains His Artificial Intelligence Grok!

//

Lerato Khumalo

Elon Musk-owned social media platform X (formerly Twitter) quietly announced a change that adds default user data to its artificial intelligence training pool called Grok. The change was noticed by platform users on Friday.

Elon Musk Teaches Data on X to Grok!

Grok is being promoted as a large language model (LLM) developed by Elon Musk’s X and intended to be a competitor to OpenAI’s ChatGPT chatbot. Grok is claimed to be less politically correct and more humorous. X users may be concerned about their information being passed on to Musk’s chatbot, and can learn more about how to turn this feature off.

Data Protection Commission Confused!

This development has caught the attention of X’s European privacy watchdog, the Irish Data Protection Commission (DPC). The DPC told TechCrunch that it was “surprised” by the platform’s move. The DPC said it is awaiting a response from X and expects further interaction early next week.

“The DPC has been engaging with X on this issue for months and our most recent interaction was yesterday,” DPC assistant commissioner Graham Doyle said. “We are therefore surprised by today’s developments. We have contacted X today and are awaiting a response.”

The Data Protection Committee is responsible for overseeing X’s compliance with the European Union’s General Data Protection Regulation (GDPR). The GDPR provides for fines for confirmed breaches of up to 4% of global annual turnover.

You May Have Given Consent Without Realizing It

The text accompanying X’s Grok data sharing setting states, “Allow us to use your posts, as well as your interactions with Grok, your input, and your results, for training and fine-tuning purposes.” The smaller, gray text states, “To continually improve your experience, we may use your posts, as well as your user interactions with Grok, your input, and your results, for training and fine-tuning purposes.” X also notes that such data “may be shared with our service provider xAI for these purposes.”

Because the language is vague, it’s unclear whether X used all user data to train Grok or just interactions with the chatbot. Either way, the company needs a valid legal basis in the EU to process people’s data in accordance with the bloc’s privacy laws. But it’s not clear whether it has one.

Meta’s plan to reuse Facebook and Instagram users’ data for AI training has been halted in Europe following GDPR complaints. The DPC is reportedly expecting more developments on Grok AI data sharing next week.

X was contacted to ask whether there was a legal basis for processing European data to train Grok, but at the time of writing the company’s press email simply responded with an automatic reply of “I’m busy right now, please check back later.”