A key industry body has warned that the government’s proposed law amendments to regulate deepfakes and AI-generated content, or synthetically generated information (SGI), are “unimplementable and would undoubtedly result in large-scale disruptions to various segments of India’s digital economy.”
IAMAI said in its submission that verifying user declarations on every post is neither technically feasible nor legally sound. (Shutterstock)
The Internet and Mobile Association of India (IAMAI) submitted its recommendations on the recently proposed amendments to the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules to the ministry of electronics and information technology (MeitY) late Thursday evening. The ministry had extended its date for public consultation from November 6 to 13 after receiving requests from multiple industry players.
IAMAI, whose members include Google, Meta, Snapchat, WhatsApp, Zomato, Groww, Airbnb, Amazon, Apple, Jio, Airtel, Netflix, Ola, and Reddit, held several consultations, and listed their concerns in an 11-page submission seen by HT.
The biggest concern large social media companies have with the proposed amendments is rule 4(1A), said a person closely associated with IAMAI. The rule requires significant social media intermediaries (SSMIs) to obtain user declarations on whether content is synthetically generated, deploy “reasonable and proportionate” technical measures to verify those declarations, and ensure that such content is clearly labelled.
IAMAI said in its submission that verifying user declarations on every post is neither technically feasible nor legally sound. It said this requirement would conflict with India’s “actual knowledge” standard and safe harbour protections, as laid out by the Supreme Court in Shreya Singhal v. Union of India. It would also slow down user experience, create privacy risks, and incentivise platforms to over-censor content to avoid penalties, said the body.
IAMAI gathered feedback from a diverse group of companies including social media companies, OTT players, advertisers, Indian startups, and food delivery companies. “Companies like these choose to submit their comments through an industry body like IAMAI so that they can ensure anonymity,” added the person quoted above.
IAMAI also pointed out that existing laws already cover the issue of SGI, and that provisions under the IT Act and the current IT Rules, 2021, already empower intermediaries to act against unlawful content, including deepfakes or impersonations. It also said the government’s proposed definition of SGI is too broad and vague.
“...‘artificially or algorithmically created’ could encompass any digital capture and processing pipeline, including the basic computational photography embedded in smartphones. Today, cameras routinely perform algorithmic sharpening, HDR tone mapping, noise reduction, and white balance. This expansive interpretation is equally problematic for text-based content,” says the submission, warning that this would impose unnecessary compliance burdens and hurt creative activity online.
“Another concern that companies have is that all of them, no matter from which industry, will have some sort of obligation. Even if their content is slightly altered, then they are worried they would come under the ambit of this proposed law and they would have to label their content as SGI,” said the person associated with IAMAI.
IAMAI also found the mandatory labelling requirement, including visible or audible watermarks covering at least 10% of the content, to be premature and impractical. It pointed out that there are no common global standards or reliable technologies yet for detecting or labelling AI-generated content. Lastly, the submission said the amendments blur the line between platforms that host third-party content (like social media sites) and those that generate content using their own AI tools. Applying the same intermediary rules to both, IAMAI warned, unfairly expands platform liability beyond what the law currently allows.
In closing, IAMAI noted that the proposed amendments contradict MeitY’s own India AI Governance Guidelines, which recommend setting up a multi-stakeholder expert committee to first develop and test global standards for content provenance and authentication. The body urged MeitY to align its approach with those guidelines.
HT reached out to MeitY officials but did not receive a response.
E-Paper
