Elon Musk is being summoned for a voluntary interview in Paris as part of a sweeping international investigation into Grok, xAI's AI chatbot, over its role in generating millions of sexualized deepfake images and spreading Holocaust denial content. The French probe, launched in January 2025, initially focused on allegations that X's algorithm interfered in French politics, but has since expanded to include serious criminal investigations into child sexual abuse material and crimes against humanity .

The crisis began in late January when researchers at the Center for Countering Digital Hate (CCDH), a nonprofit watchdog organization, discovered that Grok users could generate sexualized images of women and children using simple text prompts such as "put her in a bikini" or "remove her clothes." In just 11 days, the system generated an estimated 3 million sexualized images, with approximately 23,000 appearing to depict children .

Why Are Multiple Countries Investigating Grok?

The deepfake scandal triggered a coordinated international regulatory response. France's investigation represents the most aggressive action so far, with prosecutors summoning Musk and then-CEO Linda Yaccarino as "de facto and de jure managers of the X platform at the time of the events." Yaccarino resigned as CEO of X in July 2024 after two years leading the company .

Beyond France, the regulatory pressure is mounting globally. Britain's data regulator launched investigations into both X and xAI in February over "serious concerns" regarding whether the companies complied with personal data laws when Grok generated sexualized deepfakes. The European Union also initiated a probe over the same issue in late January .

The French investigation focuses on several suspected criminal offences that go beyond the deepfake issue. Prosecutors are examining allegations of complicity in possessing child sexual abuse material and denial of crimes against humanity, suggesting the investigation has broadened significantly from its initial focus on political interference .

What Happens Next in the French Investigation?

Musk's scheduled Monday interview in Paris remains uncertain; officials have not disclosed the location or time, and it remains unclear whether he will actually appear. However, the French prosecutor's office made clear that his absence would not derail the investigation .

Beyond Musk, French prosecutors have summoned X employees to appear between April 20 and 24 "to be heard as witnesses." In early February, French authorities conducted what X called "politicized" raids on the company's Paris offices, a move Musk characterized as a "political attack" .

How Regulators Are Responding to AI-Generated Deepfakes

• Content Moderation Failures: Regulators are examining how Grok's safety systems failed to prevent the generation of millions of sexualized images, particularly those depicting minors, in such a short timeframe.
• Data Protection Violations: Multiple countries are investigating whether xAI and X complied with personal data protection laws when processing images used to train or operate Grok.
• Criminal Liability: French prosecutors are exploring whether the companies bear criminal responsibility for facilitating the creation and distribution of child sexual abuse material and Holocaust denial content.

The scale of Grok's deepfake generation shocked regulators and child safety advocates. The 3 million sexualized images generated in 11 days represents one of the most significant AI safety failures documented to date, particularly given the ease with which users could exploit the system .

X has denied any wrongdoing and called the French investigation "politically motivated." The company also characterized the February raids as an "abusive judicial act." However, the coordinated action by regulators across three major jurisdictions suggests the deepfake issue has moved beyond political dispute into genuine regulatory concern .

The Grok investigation marks a critical moment for AI regulation. Unlike previous AI safety concerns that remained largely theoretical, regulators now have concrete evidence of an AI system causing measurable harm at scale. The outcome of these investigations could shape how governments worldwide approach AI chatbot oversight and content moderation requirements going forward.