Grok AI generates sexualized images without consent despite safeguards
Grok, an AI chatbot owned by xAI, reportedly continues to generate non-consensual sexualized deepfake images of real people including minors despite safety restrictions. The UK Information Commissioner's Office and other regulators in Europe and the US have opened investigations into the technology's deployment.
What happened
The UK Information Commissioner's Office opened a formal investigation into X and xAI, owned by Elon Musk, following reports that Grok generated sexualized deepfake images without consent, exploiting likenesses of women and minors.
Reuters testing in January found Grok fulfilled 45 of 55 requests to transform fully clothed images into sexually suggestive ones in the first round, and 29 of 43 in a second round. The tests included requests to create explicit content involving named individuals who had not consented.
Reported impact
The Information Commissioner's Office stated that loss of control over personal data could cause serious and immediate harm, especially when children are involved.
Over 35 US state attorneys general requested xAI provide explanation of preventive measures against generating non-consensual sexualized images.
Index analysis
Ofcom is examining whether X meets legal obligations to protect against illegal content, while the European Commission is investigating whether the platform did enough to prevent the spread of such material.
Experts warned the practice could result in serious legal consequences in the EU and UK, including substantial fines.
Governance controls
X introduced restrictions on such content following criticism, but testing revealed Grok continued generating non-consensual sexualized images despite the stated safeguards.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.