Malaysia has temporarily blocked access to Grok, an artificial intelligence chatbot developed for the social media platform X, following repeated misuse involving explicit and harmful content.
The move was announced by the Malaysian Communications and Multimedia Commission (MCMC), which said the suspension took effect on 11 January as a preventive measure while regulatory and legal processes continue.
According to the commission, Grok had been repeatedly used to generate sexually explicit material, extreme indecent content, and manipulated images created without consent. MCMC said the issue was particularly serious as some of the content involved women and children.
Grok is developed by xAI, and was introduced as part of X’s push into generative AI. In August last year, the chatbot added an image generation feature known as Grok Imagine. Through a setting labelled “Spicy Mode”, users were able to generate Not Safe for Work (NSFW) content, including semi-nude images based on uploaded photos.
MCMC said it had previously held multiple regulatory engagement sessions with X Corp. and xAI LLC, and issued formal notices on 3 and 8 January. These notices required the companies to implement effective technical safeguards and content supervision mechanisms to prevent AI-generated material that could breach Malaysian law, including Section 233 of the Communications and Multimedia Act 1998.
However, the regulator said responses from X Corp. on 7 and 9 January focused mainly on user-initiated reporting systems, without adequately addressing risks arising from the design and operation of the AI tool itself. MCMC said this approach was insufficient to prevent harm.
The temporary block places Malaysia as the second country after Indonesia to restrict access to Grok over content safety concerns.
MCMC said the suspension will remain in place until effective safeguards are implemented, particularly to prevent the generation of harmful content involving women and children. The commission also reiterated that it remains open to further engagement with X Corp. and xAI LLC, provided there is demonstrable compliance with Malaysian laws.
Members of the public have been urged to report harmful online content directly to MCMC, and to lodge police reports where necessary.
International attention on Grok’s image generation capabilities intensified earlier this month after Brazilian musician Julie Yukari revealed that an AI-generated image depicting her nearly nude had circulated online. The image was reportedly created after a user prompted Grok using a photo she had shared publicly, according to reports by Reuters.
The case has since become a reference point in discussions around consent, AI image manipulation, and the limits of generative tools operating on social platforms.
