Malaysia blocks Elon Musk’s Grok AI over fake, sexualised images

Malaysia has become the second country to temporarily block access to Elon Musk’s Grok after a global outcry over the AI tool and its ability to produce fake, sexualised images.Malaysia said it would restrict access to Grok until effective safeguards were implemented, a day after similar action was taken by Indonesia.Several governments and regulators have taken action over Grok’s image tool, which is embedded in the X social media site and has provoked outrage as it allows users to manipulate images of women and children to remove their clothing and put them in sexual positions.The Musk-led company that developed Grok, xAI, said last week the ability to generate and edit images would be “limited to paying subscribers” on X. Such users have provided personal details to the company and can be identified if the function is misused.The move has done little to quell anger about Grok, however. The Malaysian Communications and Multimedia Commission (MCMC) said on Sunday it would restrict access to Grok over the AI tool’s ability “to generate obscene, sexually explicit, indecent, grossly offensive, and nonconsensual manipulated images, including content involving women and minors”.Grok also functions through a separate website and app. It was unclear whether the Malaysian and Indonesian bans applied to Grok on X, the Grok site and app, or both. The Guardian received reports from Indonesia that people were still able to use Grok via the app and via X, although one reported the app was very slow.MCMC said it had issued notices to X and xAI this month to demand the implementation of effective technical and moderation safeguards. However, the responses it received relied mostly on user-initiated reporting mechanisms and failed to address the inherent risks posed by Grok, the MCMC said, adding that it considered this insufficient to prevent harm or ensure legal compliance.On Saturday, Indonesia also temporarily blocked the chatbot, with the country’s communications and digital minister, Meutya Hafid, saying the government viewed “the practice of nonconsensual sexual deepfakes as a serious violation of human rights, dignity, and the security of citizens in the digital space”.The UK has also raised the possibility of a ban if action is not taken, while on Saturday, Australia’s prime minister, Anthony Albanese, condemned the use of generative AI to exploit or sexualise people without their consent as “abhorrent”.Across Europe, regulators and politicians have also issued warnings over recent weeks.Germany’s culture and media minister, Wolfram Weimer, called on the European Commission to take legal steps, warning of the “industrialisation of sexual harassment”.Italy’s data protection authority said that using AI tools to create explicit images of people without consent could amount to serious privacy violations and, in some cases, criminal offences.In France, government ministers said this month they had referred sexually explicit Grok-generated content circulating on X to prosecutors and alerted the French media regulator Arcom. India’s IT and electronics ministry sent a formal notice to X on 2 January in relation to explicit images allegedly created through Grok, demanding the content be taken down and requiring a report on the actions being taken within 72 hours.