London. The European Union on Monday opened a formal investigation into Elon Musk’s social media platform X after its artificial intelligence chatbot, Grok, generated nonconsensual sexualized deepfake images on the platform.

European regulators also expanded a separate, ongoing probe into X’s recommendation systems after the company said it would switch to Grok’s AI to determine which posts users see.

The scrutiny from Brussels follows a global backlash after Grok’s image generation and editing tools allowed users to digitally undress people, including placing women in transparent bikinis or revealing clothing. Researchers said some images appeared to include children, prompting some governments to ban the service or issue warnings.

The European Commission, the EU’s executive arm, said it is examining whether X has complied with the bloc’s digital rules requiring platforms to mitigate risks linked to illegal content, including “manipulated sexually explicit images.”

That includes material that “may amount to child sexual abuse material,” the commission said, adding that such risks have now “materialized,” exposing EU citizens to “serious harm.”

Regulators will assess whether Grok meets its obligations under the Digital Services Act, the EU’s sweeping rule book aimed at protecting users from harmful online content and products.

In response to a request for comment, an X spokeswoman referred The Associated Press to an earlier statement saying the company remains “committed to making X a safe platform for everyone” and has “zero tolerance” for child sexual exploitation, nonconsensual nudity and unwanted sexual content.

In a statement dated Jan. 14, X also said it would stop allowing users to depict people in “bikinis, underwear or other revealing attire,” but only in jurisdictions where such content is illegal.

“Non-consensual sexual deepfakes of women and children are a violent, unacceptable form of degradation,” said Henna Virkkunen, an executive vice president of the European Commission.

“With this investigation, we will determine whether X has met its legal obligations under the DSA, or whether it treated the rights of European citizens — including women and children — as collateral damage of its service,” said Virkkunen, who oversees technology sovereignty, security and democracy.

Musk’s artificial intelligence company xAI launched Grok’s image tool last summer, but concerns escalated late last month when the chatbot appeared to approve a surge of requests to modify images posted by other users. The issue was amplified by Musk’s promotion of Grok as a less restricted alternative to rival chatbots and by the public visibility of its responses on X.

The EU investigation covers only Grok’s integration on X, not its website or standalone app, because the DSA applies only to the bloc’s largest online platforms.

There is no deadline to resolve the case. It could end with X committing to changes or facing a substantial fine.

In December, Brussels fined X 120 million euros (then about $140 million) under the earlier DSA investigation, citing shortcomings including blue checkmarks that violated rules against deceptive design practices and increased risks of scams and manipulation.

The EU has also questioned X over allegations that Grok generated antisemitic content and has requested additional information.

Malaysia and Indonesia temporarily blocked access to Grok earlier this month in response to the controversy. Malaysian authorities lifted the restriction on Friday after X implemented additional security and preventive measures, though officials provided few details and said monitoring would continue.