Indian government to Elon Musk’s X: 'Mind Your' Grok AI chatbot

Elon Musk's Grok AI model, integrated into the social media platform X (formerly known Twitter), has come under scrutiny from India’s Ministry of Electronics and Information Technology (MeitY) for generating inflammatory content. The AI bot has reportedly made abusive remarks in Hindi and controversial comments about political figures, including Prime Minister Narendra Modi and Rahul Gandhi, raising concerns over content moderation practices.
As reported by Economic Times, a senior government official confirmed that MeitY is actively engaging with X to address the issue. "The government has taken note of this, and we are deeply engaged with X on this issue. We are hoping for speedy action," the official told ET. The ministry is considering whether X and its AI model, Grok, have violated the IT Intermediary Rules 2021, which could render them liable for legal action.
Since Sunday (March16), Grok’s controversial posts have gone viral, amassing up to 80,000 views as users bait the bot with provocative prompts. The exchange began when a user asked Grok to identify their "10 best mutuals," received no immediate response, and then addressed the AI with Hindi slang.In response, Grok replied with "Oi bhos****la, chill kar," before providing the requested list of mutual followers.

While some responses have been described as humorous, experts warn that misuse of AI for hate speech is both unlawful and harmful.
Cybersecurity lawyer Prashant Mali highlighted that Section 79(1) of the IT Act provides intermediaries like X with immunity for third-party content. However, this immunity is conditional under Sections 79(2) and (3), which require intermediaries to act passively and avoid unlawful activities. Mali suggested that MeitY could revoke X’s safe harbor provisions and file a formal complaint.
Experts have also criticised Grok for failing to meet basic safety standards, despite being hailed as a high-performing AI model. The bot’s "rebellious streak," as described in a 2023 blog post by xAI, has raised questions about the balance between free speech and responsible AI usage.