India orders Musk’s X to fix Grok over ‘obscene’ AI content

India has ordered Elon Musk’s X to make quick technical and procedural adjustments to its AI chatbot Grok after customers and lawmakers flagged the technology of “obscene” content material, together with AI-altered photos of girls created utilizing the software.

On Friday, India’s IT ministry issued the order directing Musk’s X to take corrective motion on Grok, together with proscribing the technology of content material involving “nudity, sexualization, sexually express, or in any other case illegal” materials. The ministry additionally gave the social media platform 72 hours to submit an action-taken report detailing the steps it has taken to stop the internet hosting or dissemination of content material deemed “obscene, pornographic, vulgar, indecent, sexually express, pedophilic, or in any other case prohibited underneath legislation.”

The order, reviewed by Trendster, warned that failure to conform might jeopardize X’s “secure harbor” protections — authorized immunity from legal responsibility for user-generated content material underneath Indian legislation.

India’s transfer follows considerations raised by customers who shared examples of Grok being prompted to change photos of people — primarily girls — to make them seem like sporting bikinis, prompting a proper grievance from Indian parliamentarian Priyanka Chaturvedi. Individually, current studies flagged situations wherein the AI chatbot generated sexualized photos involving minors, a problem X acknowledged earlier on Friday was attributable to lapses in safeguards. These photos have been later taken down.

Nonetheless, photos generated utilizing Grok that made girls seem like sporting bikinis by means of AI alteration remained accessible on X on the time of publication, Trendster discovered.

The newest order comes days after the Indian IT ministry issued a broader advisory on Monday, which was additionally reviewed by Trendster, to social media platforms, reminding them that compliance with native legal guidelines governing obscene and sexually express content material is a prerequisite for retaining authorized immunity from legal responsibility for user-generated materials. The advisory urged corporations to strengthen inside safeguards and warned that failure to take action might invite authorized motion underneath India’s IT and prison legal guidelines.

“It’s reiterated that non-compliance with the above necessities shall be considered critically and will lead to strict authorized penalties in opposition to your platform, its accountable officers and the customers on the platform who violate the legislation, with none additional discover,” the order warned.

The Indian authorities stated noncompliance might result in motion in opposition to X underneath India’s IT legislation and prison statutes.

India, one of many world’s greatest digital markets, has emerged as a vital check case for the way far governments are keen to go in holding platforms accountable for AI-generated content material. Any tightening of enforcement within the nation might have ripple results for international expertise corporations working throughout a number of jurisdictions.

The order comes as Musk’s X continues to problem facets of India’s content material regulation guidelines in court docket, arguing that federal authorities takedown powers danger overreach, even because the platform has complied with a majority of blocking directives. On the identical time, Grok has been more and more utilized by X customers for real-time fact-checking and commentary on information occasions, making its outputs extra seen — and extra politically delicate — than these of stand-alone AI instruments.

X and xAI didn’t instantly reply to requests for touch upon the Indian authorities’s order.