Malaysia and Indonesia block Elon Musk’s Grok due to obscene, non-consensual content
Nikolas Kokovlis | Nurfoto | Getty Images
Malaysia and Indonesia blocked access to Elon Musk’s AI chatbot Grok over the weekend over concerns that the tool was being used to create non-consensual, suggestive and suggestive content.
Malaysian regulators ordered Temporary restrictions will be placed on xAI’s chatbot on Sunday following “repeated errors by X Corp” to address content risks associated with the AI tool.
The move comes just a day after Indonesia stepped in to temporarily deny access to Grok over similar concerns and asked X officials to clarify the issue. According to CNBC’s translation of the statement.
The Southeast Asian countries’ actions follow the discovery that some users of the AI tool were producing non-consensual sexually explicit images and deepfakes, including depictions of scantily clad minors.
Musk’s company recently updated its Grok Imagine features, providing greater reach by making it easier to create images from text-based prompts on the platform, which integrates with Musk’s social media platform X.
At a time when concerns are growing about Grok’s content moderation policy, xAI announced will limit image creation and editing features to paying subscribers in an effort to close security loopholes that allow sexualized output.
Musk, responds About X also suggested that users who create illegal content through Grok would face consequences equivalent to uploading such material directly to the social media platform.
CNBC attempted to reach xAI for comment on the weekend’s developments. An automated message returned in the company’s press email saying “Old Media Lies.”
“Inadequate” responses
But amid the controversy, X’s public and private responses failed to assuage concerns from regulators in Indonesia and Malaysia, as well as others who launched investigations.
The Malaysian Communications and Multimedia Commission said X’s “inadequate” responses “relied primarily on user-initiated reporting mechanisms and failed to address inherent risks arising from the design and operation of the AI tool.”
“Accordingly, the restriction is being implemented as a preventive and proportionate measure while legal and regulatory processes are ongoing,” the watchdog said. he said. “Access to Grok will remain limited until effective measures are taken to block content, especially those featuring women and children.”
Both Indonesia and Malaysia have strict anti-pornography laws that prohibit explicit and sexual content from being shared more widely online.
Meanwhile, Meutya Hafid, Indonesia’s Ministry of Communications and Digital Affairs, said: “The government considers non-consensual sexual deepfakes as a serious violation of human rights, dignity and security of citizens in the digital space.”
Hafid also classified the misuse of artificial intelligence to create fake pornography as a form of “digital-based violence,” according to statements shared by state-owned Antara News.
Authorities in other jurisdictions, including the European Union, the United Kingdom, Brazil and India, also He called for an investigation into Grok’s role in facilitating obscene and non-consensual deepfakes.
Some Democratic lawmakers in Washington, meanwhile, have suggested that app stores suspend the AI tool at least until Musk implements major changes.
In an email to CNBC last week, the Justice Department said it “takes AI-generated child sexual abuse material extremely seriously and will aggressively prosecute any producer or owner of such content.” [child sexual abuse materials]”
“We continue to explore ways to optimize practices in this area to protect children and hold accountable individuals who use technology to harm our most vulnerable,” he added.
— CNBC’s Lora Kolodny contributed to this report.
