Malaysia has become the second country to temporarily block access to Elon Musk’s AI chatbot Grok following global concern over its ability to generate fake, sexualised images.
The restriction was announced on Sunday, a day after Indonesia took similar action. Malaysian authorities said access to the tool would remain limited until effective safeguards were put in place.
Grok’s image-generation feature, embedded in the X social media platform, has drawn widespread criticism for allowing users to manipulate images of women and children, including the removal of clothing and the creation of sexualised content without consent.
The chatbot is developed by xAI, a company owned by Musk. Last week, xAI said image generation and editing would be restricted to paying subscribers on X, who are required to provide personal details and can be identified if the feature is abused.
The move has failed to ease concerns. Malaysia’s Communications and Multimedia Commission (MCMC) said Grok was capable of producing “obscene, sexually explicit, indecent, grossly offensive, and non-consensual manipulated images”, including content involving women and minors.
Grok is also available through a standalone website and mobile application. It remains unclear whether the Malaysian and Indonesian restrictions apply to Grok on X, the separate website, the app, or all three. Reports from Indonesia suggested users could still access Grok via X and the app, although performance appeared limited.
The MCMC said it had issued notices to X and xAI earlier this month, demanding stronger technical and moderation controls. It said the responses received relied largely on user reporting systems and did not adequately address the risks posed by the tool.
Indonesia announced its own temporary block on Grok on Saturday. The country’s communications and digital minister, Meutya Hafid, said the creation of non-consensual sexual deepfakes constituted a serious violation of human rights and personal dignity.
Other governments have also raised concerns. The UK has said a ban remains possible if sufficient action is not taken, while Australia’s prime minister, Anthony Albanese, described the use of generative AI to exploit or sexualise individuals without consent as “abhorrent”.
Across Europe, regulators and politicians have issued warnings. Germany’s culture and media minister, Wolfram Weimer, urged the European Commission to take legal action, warning of the “industrialisation of sexual harassment”.
Italy’s data protection authority said the use of AI to create explicit images without consent could represent serious privacy breaches and, in some cases, criminal offences.
In France, ministers said this month they had referred explicit Grok-generated material circulating on X to prosecutors and alerted the national media regulator, Arcom.
India’s IT and electronics ministry also intervened, sending a formal notice to X on 2 January demanding the removal of explicit content allegedly created using Grok and requesting a report on actions taken within 72 hours.
