Ireland’s Data Protection Commission (DPC) has opened a fresh inquiry into X and its AI chatbot, Grok, over the creation of harmful non-consensual sexualised deepfakes based on data collected from EU citizens.

The DPC said that the purpose of its “large-scale” inquiry is to determine whether the social media platform has complied with fundamental GDPR obligations regarding the processing of personal data and data protection rules.

“The DPC has been engaging with XIUC since media reports first emerged a number of weeks ago concerning the alleged ability of X users to prompt the Grok account on X to generate sexualised images of real people, including children,” said the regulator’s deputy commissioner, Graham Doyle. 

The Irish regulator is the latest to launch an investigation into Grok’s reported generation of sexualised deepfakes, coming after similar probes kicked off from the ICO and Ofcom in the UK, as well as from regulators across Spain, Canada, California, India, and France, where prosecutors went as far as raiding X’s Parisian offices.

Despite being the lead supervisory authority for X within the EU and EEA, the DPC is also behind the EU in opening a formal investigation, with Brussels having opened an inquiry into Grok last month under the auspices of the Digital Services Act (DSA). 

Others have taken a different tack, with Indonesia and Malaysia being the first countries to block Grok entirely in January, while last week, Brazilian regulators gave X five days to stop the chatbot from generating sexualised deepfakes.

Last month, X previously announced that it would prohibit Grok from editing images of real people into “revealing clothing”, but later testing by Reuters showed these limits were not working in practice, with the bot producing sexualised imagery in response to 45 of 55 prompts, including when specifically told the subject was vulnerable or would be humiliated by its output.     

According to the Centre for Countering Digital Hate, Grok has produced an estimated 3 million sexualised images, including 23,000 that the non-profit said appear to depict children, with public figures from Taylor Swift to the Deputy Prime Minister of Sweden caught up in the flood.

Recommended reading

• Grok Bot Publishes 370,000 User Conversations on Google
• xAI Under Fire Over Sexualised Images from Grok
• ICO Opens Investigations Into X and xAI

But as the furore rolls on and the voices calling for Grok, and even X itself, to be banned outright grow louder, new data shows that the platform might be benefiting from the negative attention. 

Figures from research firm Apptopia show that over January, Grok’s share of the US chatbot market jumped to 15.2%, its highest point yet and up 3% on December’s figure. That puts X’s wayward chatbot behind just Gemini and ChatGPT, and ahead of both Copilot and Claude.

Likewise, analysis by SimilarWeb found that Grok’s popularity has increased since the new year, with the bot seeing 314 million visits over January, compared to 271.1 million in December, making it one of the 100 most-visited websites in the world.