bbieron@platformeconomyinsights.com

Brazil Orders Grok to Stop Sexualized Images

Feb 21, 2026

Report from DigWatch

In Brief – Brazil has ordered X to stop AI chatbot Grok from generating sexualized images, adding to the global regulatory pressure on Elon Musk’s social media platform and AI chatbot. The directive, issued by Brazil’s National Data Protection Agency, chief prosecutor, and National Consumer Rights Bureau, requires X to implement immediate safeguards preventing the creation of sexualized content involving children, adolescents, or non-consenting adults. Brazilian authorities said that despite X claims that it deleted thousands of posts and suspending hundreds of accounts after an earlier warning, users were still able to produce sexualized deepfakes through Grok. Officials also criticized the company for lacking transparency in its response. Elon Musk and X have had a fraught relationship with Brazilian authorities, including a lengthy standoff last year with Supreme Court Justice Alexandre de Moraes over a court-ordered crackdown on populist political activism online, which temporarily led to X being shut down in the country before the platform backed down.

Context – Last summer, xAI launched Grok Imagine for paid subscribers. It included a “spicy mode” that permitted the creation of sexually suggestive content, including partial nudity, which was on brand with X’s more permissive content moderation. The image tool quickly generated complaints from civil society and child safety advocates but there was little public notice until Reuters reported on what it called a “flood of nearly nude images of real people” on Grok in markets around the world. Although Musk initially dismissed the charge, the company eventually responded to regulatory pressure with what Elon Musk described as a local law policy, with the AI chatbot capable of generating only images that he said are legal in each region, which in the US he claimed met an “R rating” and were “not real” people. The changes did not satisfy critics who claimed “sexualized” images could still be created by Grok, especially by the standalone Grok app that was not part of X. Defining “sexualized” and “intimate” or whether an image looks like a real person seems easy enough everywhere.

View By Monthly
Latest Blog
India Adds Deepfakes to Tightened Social Media Takedown Regime

Report from MediaNews4U In Brief – India has amended its rules requiring social media platforms such as Meta, YouTube and X to remove unlawful material, adding AI-generated “deepfakes” to the regime. The new rules, which go into effect on February 20, bring...

French Retailers and Government Backers Resisting Online Marketplaces

Report from Reuters In Brief – France’s Minister of Small and Medium Businesses says Shein and other online retail platforms will face a “year of resistance" from traditional storefront retailers and their backers in the government. Serge Papin argued that retail...

Federal Jury Finds Uber Financially Liable for Driver Sexual Assault

Report from Bloomberg In Brief – A federal court jury in Arizona has ordered Uber to pay $8.5 million in damages to a young woman who was allegedly raped by her driver in 2023. The decision marks the first time Uber, which has been fighting such claims for a decade,...

EU Commission Directs Google to Play Nice with Chatbot Competitors

Report from Courthouse News In Brief – The European Commission has announced two “specification proceedings” to direct Google in setting policies related to third-party AI chatbot competitors that comply with its Digital Markets Act (DMA) obligations. One process...

Platform Economy Insights produces a short email four times a week that reviews two top stories with concise analysis. It is the best way to keep on top of the news you should know. Sign up for this free email here.

* indicates required