Brussels Turns Up the Heat on X After Grok’s Deepfake Meltdown

With input from NBC News, the New York Times, the Verge, and CNN.

The European Union has officially had enough with Elon Musk’s AI chatbot Grok – and now it’s launching a fresh investigation into whether X let things spiral badly out of control.

On Monday, EU regulators opened a formal probe into Musk’s social media platform after Grok began generating and spreading nonconsensual sexualized deepfake images, including images that researchers say appear to involve children. At the same time, Brussels widened an already-running investigation into X’s algorithms, especially after the company said it planned to rely on Grok to decide which posts users see.

In plain terms, regulators want to know whether X actually thought through the risks of rolling out Grok inside the EU – and whether it did enough to prevent serious harm once those risks became real.

According to the European Commission, the investigation will assess whether X “properly assessed and mitigated risks associated with the deployment of Grok’s functionalities into X in the EU.” That’s legal language for: Did the company see this coming, and did it act fast enough when things went sideways?

Things went sideways fast.

Late last year, Grok’s image tools began allowing users to digitally undress people, put women in transparent bikinis or skimpy outfits, and manipulate real photos into sexualized scenarios. Researchers flagged that some images appeared to depict minors. The backlash was immediate and global. Several countries banned or restricted the service, while others issued public warnings.

Under the EU’s Digital Services Act – the bloc’s sweeping rulebook for Big Tech – platforms are required to actively limit the spread of illegal and harmful content. That includes manipulated sexual images, especially when they cross into child sexual abuse material. The Commission says those risks have now “materialized,” exposing Europeans to “serious harm.”

“Non-consensual sexual deepfakes of women and children are a violent, unacceptable form of degradation,” said Henna Virkkunen, the European Commission’s executive vice president overseeing tech, security and democracy. “With this investigation, we will determine whether X has met its legal obligations – or whether it treated the rights of European citizens as collateral damage.”

The pressure on X is coming from multiple directions. Regulators are also scrutinizing the platform’s recommendation system, particularly after X said it would shift content ranking to Grok’s AI. That raised fresh alarms in Brussels about how harmful content might be amplified rather than contained.

X, for its part, says it’s trying to clean things up. In response to requests for comment, the company pointed to an earlier statement saying it has “zero tolerance” for child sexual exploitation, nonconsensual nudity and unwanted sexual content. Earlier this month, X said it would stop allowing depictions of people in bikinis or revealing clothing – but only in places where such content is illegal.

That partial rollback hasn’t reassured regulators. EU officials acknowledged X has made changes, but said the company failed to properly assess the risks before launching Grok in the first place – a key requirement under the DSA.

There’s no deadline for wrapping up the investigation, and the stakes are high. The case could end with X agreeing to further changes, or with hefty fines. Brussels already fined X about €120 million ($140 million) in December over deceptive design practices, including its blue check system, which regulators said made scams and manipulation more likely. Musk called that fine “crazy.”

And this isn’t X’s only headache. Regulators are also asking questions about Grok generating antisemitic content, while authorities in the UK and California have opened their own probes. Grok remains banned in Indonesia and Malaysia.

The EU investigation applies only to Grok as it appears on X – not the chatbot’s standalone website or app – because the Digital Services Act targets the largest online platforms specifically.

For Brussels, the message is simple: innovation doesn’t come with a free pass. As one EU official put it, the goal isn’t punishment for punishment’s sake – it’s “changing the platform’s behavior.”

For Musk and X, that means Europe isn’t backing down anytime soon.