Grok generated 87% of files in deepfake attacks, 1 in 6 involved children
Elon Musk's Grok AI tool has been linked to the vast majority of deepfake attacks in the first half of 2026. According to a report from Resemble AI, the model accounted for 87% of all generated files connected to such incidents.
Researchers examined 1,760 news reports and identified 821 separate attacks affecting at least 15,736 documented victims. Together, those incidents involved approximately 3.46 million synthetic images, videos, and audio files. Resemble AI's data is based on publicly reported cases, so the percentage applies only to files that could be counted and attributed to a specific tool.
Even with that caveat, the figure shows how quickly a widely available AI image generator can become a tool for misuse. Of particular concern, 137 of the 821 attacks — roughly one in six — contained sexual imagery of adults or minors without their consent.
The report comes as Musk makes a major financial bet on Grok. Following SpaceX's acquisition of xAI, Musk said during a company-wide meeting that AI revenue would surpass revenue from all other products as soon as September. SpaceX reported $2.56 billion in AI revenue in its most recent quarter, placing Grok and its infrastructure at the center of the company's future. This rapid commercial growth makes questions about how the tool is used difficult to separate from its business success.
Earlier this year, Grok faced widespread criticism after users began asking the chatbot to digitally undress people or place them in sexualized situations. Some of the generated images reportedly depicted minors, prompting investigations from regulators worldwide. The European Commission opened a formal investigation into X over sexualized images created with Grok, while California Attorney General Rob Bonta launched a probe into the creation and spread of non-consensual explicit material through the platform. Authorities in the UK, Australia, and other countries have also scrutinized the tool.
Resemble AI estimates that companies allowing or distributing the documented images could face up to $2.24 billion in potential civil liability under US law.
