Brussels, 26 January 2026, 16:15 (CET)
- The EU has kicked off a formal Digital Services Act probe into X, zeroing in on Grok’s image features and the dangers posed by illegal sexual deepfakes
- The Commission has widened a separate, ongoing probe to cover X’s recommender algorithms as well
- X says it has beefed up blocks on image editing and location tracking, though regulators caution that new risks continue to surface
On Monday, the European Commission opened a new probe into Elon Musk’s social media platform X, zeroing in on the spread of manipulated, sexually explicit images tied to its Grok chatbot. Simultaneously, regulators widened an ongoing investigation into X’s recommender systems—the algorithms that decide what content users see. 1
This is urgent because Grok’s image tools make it easier to produce non-consensual sexual deepfakes—AI-generated images that exploit real people—and these can spread fast on a platform built for sharing. Brussels is relying on the Digital Services Act, or DSA, its regulatory rulebook for big online platforms, to turn risk controls from optional steps into mandatory requirements.
EU tech chief Henna Virkkunen condemned non-consensual sexual deepfakes of women and children as “a violent, unacceptable form of degradation.” She said the investigation will look into whether X treated citizens’ rights as “collateral damage.” Reuters noted the probe could spark a reaction from the administration of U.S. President Donald Trump, especially since EU tech enforcement has already drawn criticism and tariff threats. 2
The Commission said it will probe whether X properly assessed and tackled “systemic risks” linked to Grok’s rollout in the EU, focusing on worries about manipulated sexually explicit images that might count as child sexual abuse material. It’s also calling for a one-time, special risk assessment report on Grok features that notably changed X’s risk landscape.
X pointed to a Jan. 14 statement from owner xAI, which said it had restricted image editing for Grok users and blocked generating images of people in revealing clothing where such content is illegal—though it didn’t name specific countries. An X spokeswoman also cited an earlier company statement stressing X’s commitment to safety, with “zero tolerance” for child sexual exploitation and non-consensual nudity. The Commission has yet to set a deadline for resolving the case. 3
The dispute has gone beyond Europe. Earlier this month, Indonesia, the Philippines, and Malaysia all temporarily blocked Grok. But after xAI rolled out safety measures, the Philippines and Malaysia decided to lift their bans.
Two weeks ago, Britain’s media watchdog Ofcom opened an investigation, intensifying pressure on X as regulators assess whether platforms are doing enough to stop the spread of illegal sexualised deepfakes. 4
Earlier this month, the Commission blasted AI-generated images of nude women and children spreading on X, calling them “unlawful” and “appalling.” This sharp condemnation mirrors concerns voiced by regulators around the globe.
Irish MEP Regina Doherty highlighted that the case exposed shortcomings in how authorities keep tabs on fast-changing AI technologies. “The AI Act must remain a living piece of legislation,” she argued, emphasizing that EU rules have to be enforceable “in real time when serious harms occur.”
The Grok case comes amid a ramp-up in DSA enforcement hitting big platforms. The Commission has moved against Meta and TikTok too, zeroing in on transparency rules that require giving researchers access to platform data. 5
X is facing DSA penalties from Brussels. In December, the Commission hit the company with a 120 million euro fine over a “deceptive” blue checkmark, ads absent from its repository, and limited public data access for researchers. 6
The probe will likely focus on technical details: the risks X flagged before launch, the safeguards it put in place, and whether the Commission finds those steps enough. Regulators have said they can impose interim measures—temporary orders to demand changes while the investigation runs—if the fixes fall short.
The Commission’s new investigation zeroes in on Grok features embedded directly in X, excluding Grok’s standalone app or website, as the DSA targets the EU’s biggest online platforms. Officials confirmed that the case around recommender systems now covers X’s planned shift to Grok-based ranking.
