The Central government is contemplating amendments to Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021, specifically targeting deepfakes. The Ministry of Electronics and Information Technology (MeitY) is reportedly considering defining deepfakes explicitly and making it mandatory for intermediaries to exert "reasonable efforts" to prevent their hosting. This move aims to provide legal backing to two advisories issued by the ministry on deepfakes. Proposed amendments are expected to be discussed within the ministry in the upcoming week.

The initiative follows three meetings held by IT Minister Ashwini Vaishnaw and Minister of State Rajeev Chandrasekhar with social media and technology companies in November and December. MeitY had also issued two advisories on February 21 and December 26, along with two letters on November 6 and November 7, addressing the issue of deepfakes.

MeitY is considering three specific amendments. Firstly, it aims to define deepfakes as any content—audio, visual, or audio-visual—created, edited, or altered through electronic means that could be perceived as true. To safeguard against unintended consequences, the ministry is contemplating limiting the definition to certain identified harms, although the nature of these harms remains unclear at this stage. Relevant amendments to rule 3(1)(b) are under consideration to prohibit intermediaries from hosting deepfakes.

Secondly, in alignment with Chandrasekhar's discussions and public statements, the ministry is exploring the obligation for intermediaries to remind users of prohibited content under rule 3(1)(b) every 15 days using clear and precise language. Eleven categories of content, including those violating intellectual property rights, impersonating others, or posing a threat to national security, are not allowed under this rule. This proposal contrasts with the existing rule, requiring annual reminders of terms and conditions, privacy policies, and user agreements.

Lastly, the ministry is contemplating an expansion of the "grievance" definition under the IT Rules. This would allow "complaints" related to user-generated content violating rule 3(1)(b) to be directly sent to the intermediary's grievance officer. Such an expansion would mean that all content reported through in-app mechanisms would also be forwarded to the grievance officer.