Malaysia and Indonesia Block AI Chatbot Grok Over Sexually Explicit Deepfakes
Malaysia and Indonesia have become the first countries to block access to Elon Musk’s AI chatbot Grok after it was discovered that there has been widespread misuse of the tool to create sexualised and non-consensual images of women and children.
Grok, which is built into Musk’s social media platform X and also available as a standalone service, allows users to generate and edit images. In recent weeks, it has been used to manipulate photos of real people, showing them in revealing clothing or explicit poses without their consent. The trend, often described as “digital undressing”, has sparked international backlash.
Both governments announced their decisions over the weekend. Indonesia’s digital affairs minister Meutya Hafid said the move was aimed at “protect women, children and the broader public from the risks of fake pornographic content generated using artificial intelligence technology.” Malaysia followed with a temporary ban, pointing to what it called repeated misuse of Grok to create obscene and harmful material.
Malaysia’s Communications and Multimedia Commission said it had previously warned X and its AI developer xAI to tighten safeguards. But after reviewing the company’s response, the regulator said X focused mainly on user reporting tools and failed to deal with deeper design risks. “MCMC considers this insufficient to prevent harm or ensure legal compliance.”
Indonesian officials said using Grok to produce sexually explicit content violates human rights and dignity. Authorities in both countries, which have strict anti-pornography laws, said the blocks would remain until effective protections are put in place.
The bans follow growing concern worldwide. Researchers in Europe have found that a large share of Grok’s image outputs feature people in minimal clothing, while prompts frequently involve removing or altering clothing. Victims have described distress and embarrassment after discovering manipulated images of themselves circulating online.
xAI initially responded to media queries with an automated message saying “Legacy Media Lies.” X later said it takes action against illegal content and warned: “Anyone using or prompting Grok to make illegal content will suffer the same consequences as if they upload illegal content.”
Last week, Grok limited some image-generation features on X to paying subscribers, but critics said the change did little to fix the core problem. Non-subscribers can still edit images, and similar tools remain available through Grok’s own website and app.
Pressure is also building in the UK. Prime Minister Keir Starmer has described the use of Grok to create sexualised images as “disgraceful” and “disgusting”, while Technology Secretary Liz Kendall said she would support action to block X if it fails to follow online safety laws. On Monday, UK regulator Ofcom opened a formal investigation into whether X is doing enough to protect users from illegal content.
Musk has defended Grok in the past, accusing governments of wanting to suppress free speech and dismissing some criticism online. However, regulators in Europe, Asia and beyond are now watching closely as scrutiny of AI-generated deepfakes intensifies.
