The Commission said it has extended a document retention order originally sent to X last year, requiring the company to preserve records linked to its algorithms and the dissemination of illegal content until the end of 2026. The move is intended to ensure the bloc can effectively assess X’s compliance with European Union rules, particularly following criticism of Grok’s output.
Commission spokesperson Thomas Regnier said the decision reflects doubts about whether the platform is meeting its obligations under EU law. He explained that the order obliges X to keep internal documents intact so regulators can request access if necessary. However, he stressed that the step does not amount to the opening of a new formal investigation under the EU’s Digital Services Act (DSA).
The action comes as criticism of Grok spreads across Europe. Sweden on Thursday joined other countries in condemning the AI-generated images, after a user prompt reportedly produced sexualised content involving Sweden’s deputy prime minister. Swedish Prime Minister Ulf Kristersson described the images as “a kind of sexualized violence,” calling them “distasteful, unacceptable and offensive.”
Concerns have also been raised in the United Kingdom. Prime Minister Keir Starmer reiterated calls for X to take urgent action after a report by the Internet Watch Foundation (IWF) said Grok was being used to generate sexualised images of children. “It’s disgusting. And it’s not to be tolerated,” Starmer said in an interview with national radio network Greatest Hits Radio.
The IWF, a British non-profit organisation focused on combating online child sexual abuse, said it had identified criminal imagery involving children aged between 11 and 13 that appeared to have been created using Grok. Ngaire Alexander, head of the organisation’s reporting hotline, warned that such tools risk pushing sexual AI imagery of children into the mainstream, describing the trend as unacceptable.
X did not immediately respond to a request for comment. However, the company’s Safety account said earlier this week that it removes all illegal content from the platform, including child sexual abuse material, and permanently suspends accounts involved. It added that users who prompt Grok to generate illegal content face the same penalties as those who upload such material directly.
The episode underscores the growing regulatory and political scrutiny facing AI tools deployed by major online platforms, as governments seek to ensure emerging technologies do not amplify harmful or illegal content.