Statement by the Privacy Commissioner of Canada on investigation into Grok chatbot and sexualized deepfakes

Privacy Commissioner of Canada Philippe Dufresne today issued the following statement on his investigation into the Grok chatbot’s artificial intelligence-powered image-generation tool.

Artificial intelligence and other emerging technologies are fueled by the mass collection and use of data, including personal information.

While technological innovation can bring important economic, societal, and public-interest benefits, the value of innovation can be jeopardized when it is not deployed with appropriate safeguards.

A few months ago, reports emerged that Grok’s image-generation tool was being used to generate millions of sexualized deepfakes without consent.

Given the serious risks for Canadians’ fundamental right to privacy and the potential harms, I initiated an investigation on this issue into X Corp. and xAI.

The investigation has found that X Corp. and xAI violated Canada’s federal private-sector privacy law by launching the Grok AI-powered image-generation tool without implementing appropriate safeguards from the outset.

This lack of protections allowed users to create and share sexualized deepfakes, including many targeting women and children. According to researchers, Grok was at one point generating well over 6,000 sexualized images per hour.

We know that sexualized deepfakes can have devastating impacts on victims.

Canadians – and especially our children – must be able to safely navigate online spaces. Just yesterday, the Government tabled legislation that would create new requirements for social media services and AI chatbot services.

I welcome these efforts to implement measures to make the online world a safer space.

This is also why I launched the Grok investigation.

Since the Grok issue came to public attention, and during the investigation, X Corp. and xAI have implemented safeguards to reduce the risk that the tool will be misused to produce sexualized deepfakes, as well as mechanisms to review and remove this harmful content.