Users exploited Grok to generate non-consensual sexually explicit images/videos of women via prompts altering photos (e.g., "put her in a bikini" or remove clothing).
MeitY notice demands immediate removal of obscene content, technical review of Grok, and 72-hour Action Taken Report; warns of legal consequences for non-compliance.
Highlights privacy violations, potential for harassment, and gaps in AI safeguards; echoes global scrutiny including French reports on similar explicit outputs.
The Ministry of Electronics and Information Technology (MeitY) has issued a stern notice to X (formerly Twitter), demanding immediate action to curb the misuse of its AI chatbot Grok for generating and disseminating obscene, sexually explicit, and non-consensual adult content — including manipulated images and videos of women, often created through prompts that "undress" or sexualize real individuals.
The directive follows a surge in viral examples where users prompted Grok to alter photos — primarily of women — to add bikinis, remove clothing, or produce suggestive poses, leading to widespread circulation on the platform. Lawmakers and users flagged the trend as a gross violation of privacy and dignity, with reports highlighting fake accounts being used to target individuals without consent. The controversy has escalated amid global backlash, including French authorities reporting similar sexualized outputs (including of minors) to prosecutors.
In its January 2 notice, MeitY described the incidents as a "serious breakdown" in safeguards, accusing X of failing statutory due diligence under the IT Act, 2000, and IT Rules, 2021. The ministry directed X to:
Remove or disable all such unlawful content immediately
Conduct a technical and governance review of Grok's prompt-processing systems
Restrict generation of nudity, sexualization, or explicit material
Submit a detailed Action Taken Report (ATR) within 72 hours
Failure to comply could result in loss of legal immunity for intermediaries and potential penal action under cyber, criminal, and child protection laws.
The issue gained prominence after Shiv Sena (UBT) MP Priyanka Chaturvedi wrote to IT Minister Ashwini Vaishnaw, highlighting how men were using fake accounts to upload women's photos and prompt Grok to "minimise clothing" or sexualize them. Chaturvedi called it "unacceptable" and urged safeguards to protect women's safety.
X and xAI have not issued a detailed public response, though Grok itself acknowledged "lapses in safeguards" in some related cases. The platform's "Spicy Mode" — designed for more permissive adult-themed outputs — has been criticized for enabling such misuse, contrasting with stricter filters on competitors like ChatGPT or Gemini.
Experts warn that Grok's integration into X lowers barriers for creating non-consensual deepfakes and explicit content, raising concerns over harassment, revenge porn, and exploitation. The controversy underscores growing regulatory pressure on generative AI in India, especially ahead of stricter guidelines expected in 2026.
