AI-created content labelling to curb cybercrime, sexual abuse: Official

NEW DELHI: The government is working on a mandate to use a watermark on artificial intelligence or AI-generated content to prevent cybercrime and rampant misuse including deepfakes and sexually-explicit images and videos, according to a senior official.

"We have recently came out with the draft provisions with regard to watermarking of AI-generated content, which required all AI-generated content to be labelled, because very often a lot of cybercrime is done with AI-generated content, and deepfakes are created which can cause problems," Abhishek Singh, additional secretary, Ministry of Electronics & IT (MeitY), and CEO of IndiaAI Mission told ET.

The draft guidelines, according to him, could deter any AI-generated content potentially leading to law and order issues or that could drive social unrest or which could lead to child sexual abuse.


Late last year, the government came out with the draft amendments to the Information Technology (Intermediary Guidelines and Digital Media Ethics Cde) Rules, 2021, saying that it remains committed to "ensuring an open, safe, trusted and accountable Internet" for users.

On the back of public and parliamentary discussions, the draft amendment is aimed to curtail increasing proliferation of synthetically-generated information or AI-driven deepfakes causing user harm, including sexually-oriented content, electoral manipulations, misinformation and impersonation.

The draft rules aimed at strengthening due diligence obligations for intermediaries, particularly social media platforms such as X, Meta and Instagram.

"The ministry has come out with intermediary guidelines which regulate the removal of such content when brought to the notice of social media platforms. There are ample provisions in the Indian IT Act 2000, and the Bharatiya Nyaya Sanhita (BNS), 2023 which will regulate and limit the harm that misuse of AI can cause," the top official said.

The MeitY's cyber laws division, that enforces laws such as the IT Act 2000, Digital Personal Data Protection Act 2023, and the IT (Intermediary Guidelines) Rules 2021, is working on finalising the contours of the new amended guidelines.

Further, Singh said that a new framework would "come out soon."

On January 2, the ministry pulled up billionaire Elon Musk-owned microblogging platform X (formerly Twitter), and directed it to remove obscene or sexually explicit and unlawful content generated by its AI interface Grok and xAI, and warned of suitable action under the existing laws of the land.

It has also sought an action taken report (ATR) within a three-day period from the US-based X.

In a notice to X, the ministry also said, "It is reiterated that non-compliance with the above requirements shall be viewed seriously and may result in strict legal consequences against your platform, its responsible officers and the users on the platform who violate the law, without any further notice, under the IT Act, the IT Rules, the BNSS, the BNS and other applicable laws."

The development comes close on the heels of Opposition leader from Shiv Sena (UBT) Priyanka Chaurvedi writing to the Union IT minister Ashwini Vaishnaw, saying that the misuse of AI tools on social media, especially X's Grok AI feature, to sexualize women's photos without consent as "unacceptable, unethical and criminal" and urged the government to take "urgent action" to protect women's safety and dignity on digital platforms.

However, on December 29, 2025, MeitY issued an advisory to social media intermediaries reminding them about the penal provisions under various statutes, including sections 67, 67 A and 678 of the IT Act; relevant provisions of the BNS; the Indecent Representation of Women (Prohibition) Act, 1986; the Protection of Children from Sexual Offences Act, 2012; the Young Persons (Harmful Publications) Act, 1956; and other applicable laws, under which the hosting, displaying, uploading, publication, transmission, sharing of obscene, pornographic, paedophilic, indecent, vulgar, or sexually-explicit content constitutes a punishable offence.

On Wednesday, Texas-headquartered X responded to the ministry, outlining its content takedown policies on non-consensual sexual images but failed to take any specific action.

However, the government, according to the sources, has further sought response including particular takedowns by the digital platform as well as preventive steps for any such violation in future.

Meanwhile, the UK's regulator for the communications services Ofcom, has also contacted with Musk's company xAI following reports its AI tool Grok making sexual visuals of children and women, and is currently probing the matter.