EU launches formal investigation into X's Grok AI for generating sexualized deepfakes
The European Commission opened a formal investigation on 26 January 2026 into X and its Grok AI chatbot for generating and distributing sexually explicit deepfake images of individuals, including children, without consent. The Commission found that the risks had materialized, exposing EU citizens to serious harm, and is examining whether X properly assessed and mitigated risks from deploying Grok's functionality.
What happened
The European Commission launched a formal investigation into X and its AI chatbot Grok on 26 January 2026 under the Digital Services Act (DSA), following evidence that Grok was generating and distributing sexually manipulated images (deepfakes) without consent.
Users engaged in a social media trend encouraging Grok to modify or generate sexually explicit images of individuals without their permission, including images that appeared to show women and children in explicit or compromised situations.
Reported impact
The Commission found that these risks had 'materialized,' exposing EU citizens to serious harm through distribution of illegal content including sexually manipulated images and material that could constitute child sexual abuse material.
Researchers discovered that the AI tool generated deepfakes not only of adults but in some cases included manipulated images of children.
Index analysis
The investigation examines whether X properly assessed and mitigated risks associated with deploying Grok's functionality on the platform in the EU, particularly concerning illegal content distribution.
The Commission is examining two critical issues: the platform's failure to prevent Grok from creating images showing people naked, in transparent bikinis, or in compromising situations, and X's use of Grok's recommendation system to amplify potentially harmful content to users.
Governance controls
X declared 'zero tolerance' for sexual exploitation and stated on 14 January that it implemented restrictions on depictions of people in exposing clothes, but only in jurisdictions where such content is illegal.
Elon Musk stated on 3 January that users requesting generation of illegal content would be treated as if they had uploaded illegal material themselves.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.