BRUSSELS, Jan 8 (Reuters) - The European Commission has ordered Elon Musk's X to retain all documents relating to its AI chatbot, Grok, for longer while the bloc ensures compliance with its rules after condemning it for producing sexualised images, a spokesperson said on Thursday.

Sweden on Thursday joined the chorus of criticism saying the AI-generated images were unacceptable, after the country's deputy prime minister was targeted by a Grok user's prompt this week.

The Commission has now decided to extend a retention order sent to X last year, which related to algorithms and dissemination of illegal content, prolonging it to the end of 2026, spokesperson Thomas Regnier told reporters on Thursday.

"This is saying to a platform, keep your internal documents, don't get rid of them, because we have doubts about your compliance ... and we need to be able to have access to them if we request it explicitly," Regnier said.

SWEDISH PM SAYS IMAGES ARE UNACCEPTABLE

He said the move did not mean the Commission had opened a new formal investigation based on the European Union's Digital Services Act, which requires online platforms to do more to tackle illegal and harmful content.

X did not immediately respond to a request for comment when contacted by Reuters.

X's Safety account said on Sunday that it removes all illegal content, including child sexual abuse material, on the platform and permanently suspended accounts involved. It said anyone using or prompting Grok to make illegal content would suffer the same consequences as if they uploaded illegal content.

Swedish Prime Minister Ulf Kristersson described the images as "a kind of sexualized violence" and said: "It's distasteful, unacceptable, offensive."

British Prime Minister Keir Starmer also called again on X to take urgent action, after a report from the Internet Watch Foundation said Grok was being used to generate sexualised images of children.

"It’s disgusting. And it’s not to be tolerated," Starmer told national radio network Greatest Hits Radio.

The Internet Watch Foundation, a British non-profit organisation focused on eradicating online child sexual abuse, said it had found criminal imagery of children aged between 11 and 13, which seemed to have been created by the use of Grok.

"Tools like Grok now risk bringing sexual AI imagery of children into the mainstream," Ngaire Alexander, head of the reporting hotline at the Internet Watch Foundation, said in a statement. "That is unacceptable."

(Reporting by Louise Breusch Rasmussen in Paris, Johan Ahlander in Stockholm and Sam Tabahriti in London; Editing by Bart Meijer and Alison Williams)