OTTAWA — Canada’s privacy commissioner says sexual deepfakes generated by Elon Musk’s Grok artificial intelligence chatbot violated Canadian privacy law and were produced without adequate safeguards to protect Canadians.Privacy Commissioner Philippe Dufresne released the findings of an investigation Thursday into Grok’s AI image-generation tools, concluding the technology was launched without sufficient consideration of privacy risks and consent requirements.The investigation was launched in January following a surge of sexualized deepfake images generated through Grok and distributed on the social media platform X.Dufresne said reports at the time suggested the system was producing approximately 6,000 sexualized images per hour, contributing to what he described as a significant privacy concern.The probe examined whether companies behind the technology obtained valid consent to collect, use and disclose personal information in the creation of deepfakes, including explicit images.Dufresne concluded the companies failed to comply with Canadian privacy law..The case attracted international attention, with regulators in the United Kingdom, the European Union and California also examining the proliferation of AI-generated sexual content.The findings come a day after the Carney government introduced Bill C-34, the Safe Social Media Act, which would establish new rules for social media platforms and AI chatbot services, including requirements aimed at protecting children online.Asked by the Western Standard whether any particular age-verification technology struck the right balance between privacy and effectiveness, Dufresne declined to endorse a specific method.“We have not endorsed a specific technology or a specific methodology,” he said. “We’re open to different approaches, but we have highlighted some of the key principles.”Dufresne said those principles include proportionality, data minimization, limits on data retention, and avoiding secondary uses of personal information.“We talk about avoiding tracking,” he said. “All of those things are key.”He added that his office will review the government’s newly introduced legislation and recommend changes if necessary.“We’re going to review to make sure if there’s some gaps we’ll make some recommendations to the appropriate parliamentary committee,” Dufresne said.The commissioner said the debate over online age assurance is taking place globally and Canada will continue to participate in those discussions.“We’re going to continue to take part of it,” he said.
