Malaysia and Indonesia Block Elon Musk’s Grok AI Chatbot
KUALA LUMPUR, Malaysia — Malaysia and Indonesia have taken the step of blocking Grok, the AI chatbot developed by Elon Musk’s xAI. This action comes amid rising concerns from global authorities about its potential misuse in creating explicit and nonconsensual imagery.
The decision highlights the increasing scrutiny over generative AI technologies capable of producing hyper-realistic images, sounds, and text. There have been worries that current safeguards are inadequate to prevent their misuse. Critics have pointed out Grok’s ability to generate manipulated images that depict women in provocative scenarios, as well as concerning images involving children.
Recently, Grok attempted to address these issues by limiting image generation to paying customers after significant backlash over sexualized deepfakes, but many argue that this response falls short of solving the core problems.
When contacted by The Associated Press for a comment, xAI’s reply was merely an automated message stating, “Legacy Media Lies.” This was similar to responses received during inquiries about the backlash against the chatbot.
The regulators in both Malaysia and Indonesia have expressed that existing controls do not effectively prevent the creation and spread of fake pornographic images, especially those involving women and minors. Indonesia’s government imposed a temporary ban on Grok, which Malaysia followed shortly after.
Indonesian Communication and Digital Affairs Minister Meutya Hafid emphasized in a statement that they view nonconsensual sexual deepfakes as a serious infringement on human rights and citizen safety within the digital landscape. The ministry aims to safeguard women, children, and the community at large from the dangers of AI-generated fake pornographic imagery.
Initial assessments revealed that Grok lacks robust mechanisms to prevent users from producing and distributing explicit content using real images of Indonesian individuals, according to Alexander Sabar, director-general for digital space oversight. Manipulated photographs, particularly those shared without consent, pose risks to privacy rights and can lead to psychological and social harm.
In Malaysia, the Communications and Multimedia Commission issued a temporary restriction on Grok due to, as they put it, “repeated misuse” of the platform for generating obscene and explicit content involving minors and women. Notices sent to xAI and X Corp. demanding improved safeguards reportedly received overly simplistic responses, primarily emphasizing user reporting.
The commission stated the restriction is a precautionary measure while legal and regulatory reviews proceed and will remain in effect until effective safeguards are established.
Investigation Launched in the U.K.
Initially launched in 2023, Grok is freely available on the X platform, allowing users to ask questions and engage with content. The image generator feature, Grok Imagine, which includes an adult content mode, was added last summer.
The actions taken by Southeast Asian countries come as the chatbot faces increasing scrutiny in regions like the European Union, the U.K., India, and France. On Monday, the U.K.’s media regulator initiated a formal investigation to determine whether Grok has met its obligation to protect citizens from illegal content.
Ofcom, the U.K. regulator, has noted that the Grok-generated images, which may depict children or sexually explicit scenarios, could be classified as pornography or child sexual abuse material. Technology Secretary Liz Kendall remarked on the disturbing nature of the content generated and shared through Grok in recent times.
