EU Moves to Ban AI-Generated Sexual Deepfakes in Wake of Grok Scandal

EU Backs Landmark Ban on AI-Generated Sexual Deepfakes

European Union member states have formally backed a prohibition on artificial intelligence systems that generate non-consensual sexual imagery and child sexual abuse material, marking one of the most significant regulatory actions against AI misuse to date.

EU ambassadors agreed on Friday to ban practices involving the generation of non-consensual sexual and intimate content, with Cyprus, which currently holds the EU's rotating presidency, confirming the decision.

From Scandal to Legislation

The ban emerged directly from one of the most high-profile AI controversies in recent memory. In late December 2025, xAI updated its Grok chatbot on Elon Musk's platform X with an image-editing feature that users quickly exploited to generate realistic sexualised images of real women and girls without their consent. The Center for Countering Digital Hate reported that Grok produced an estimated one million sexual images within just eleven days, with some depicting minors.

The European Commission responded in January 2026 by opening a formal investigation into X under the Digital Services Act, examining whether the platform had properly assessed and mitigated risks. National investigations followed in the United Kingdom, Ireland, France, and Spain.

Legal Gap Exposed

The Commission confirmed that the AI Act, as originally written, did not explicitly ban AI systems capable of generating such material, a legal gap that provided the political momentum for the new amendment. The proposed ban covers AI systems that alter, manipulate, or artificially generate realistic images or videos depicting sexually explicit activities or intimate parts of an identifiable person without their consent.

What Happens Next

The ambassadors' agreement followed a political deal among European Parliament lawmakers struck on March eleventh, which included the deepfakes prohibition as part of a broader package of amendments to the AI Act known as the Omnibus. Parliamentary committees are scheduled to vote on the measure next Wednesday, after which both bodies must negotiate a joint text. The broader discussions are expected to take approximately a year before changes can be fully implemented.