Elon Musk’s AI chatbot, Grok, is facing serious trouble in India after a trend involving “bikini” and undressing prompts went viral online. What started as people experimenting with the chatbot quickly turned into a major concern for authorities, as Grok was used to create sexualized and altered images of women and in some cases, even minors.
On January 2, India’s IT ministry stepped in and gave X, the platform that owns Grok, a strict 72-hour deadline. The government demanded that X fix its AI systems, put proper safety controls in place, and submit a detailed report explaining what actions it has taken. Officials warned that if X fails to comply, it could lose its legal protection in India protection that currently shields social media companies from being held responsible for content created by users. That deadline ends today, and so far, there’s no public word on whether X has responded.
The issue gained momentum after reports showed X being flooded with explicit images generated using Grok. Users reportedly uploaded normal photos of women and then asked the chatbot to edit them with commands like “put her in a bikini” or “remove her clothes.” These images spread quickly, raising alarms about privacy, consent, and child safety.
India isn’t the only country paying attention. France and Malaysia have also started investigations into Grok’s image generation practices. On January 1, Grok itself posted an apology after creating and sharing a sexualized AI image of two young girls, estimated to be between 12 and 16 years old, based on a user request.
A few days later, on January 4, Elon Musk and X’s safety team released statements saying users are responsible for creating illegal content. They added that X will remove such material, permanently ban accounts involved, and cooperate with local governments and law enforcement when needed.
Experts say this problem goes far beyond one chatbot or one company. AI tools that create fake or sexual images without consent are becoming more common and harder to control. A 2025 study of nearly 30 so-called “undressing apps” found that these tools fuel widespread abuse, especially against women, by turning exploitation into a product.
Governments around the world are trying to catch up. In the U.S., a new law passed in May 2025 makes it illegal to share non-consensual intimate images and AI-made deepfakes. The U.K. is considering tough penalties and even banning nudification apps altogether. China has suggested adding watermarks to AI-generated content, while Australia has urged schools to treat such incidents as serious crimes and report them to police.
Still, many experts say these steps aren’t enough. As futurist Bernard Marr wrote last year, while countries are trying to regulate AI abuse, progress has been uneven and the technology is moving faster than the rules meant to control it.
