MeitY Issues AI Advisory for Large Platforms

In a notable development regarding the regulation of Artificial Intelligence (“AI”), the Government of India released an advisory mandating platforms seeking to deploy any unreliable AI models, software or algorithms on the Indian Internet to first obtain explicit permission from the Ministry of Electronics and Information Technology (“MeitY”). Furthermore, intermediaries or platforms were strongly encouraged to ensure that their systems did not perpetuate bias, discrimination or compromise the integrity of the electoral process. Additionally, all synthetically generated media and text were to be appropriately labelled with unique identifiers or metadata for ease of identification.

To fully grasp the significance of this latest advisory, it is essential to consider a prior directive issued in November 2023. This earlier advisory instructed significant social media intermediaries, such as Facebook, Instagram and YouTube to swiftly remove deep fake content upon notification from affected individuals. Platforms complying with these takedown requests were granted safe harbour protections under Section 79(1) of the Information Technology Act, 2000.

Both advisories were issued under Rule 3(1)(b) of the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021 (“IT Rules, 2021”). This rule mandates platforms to make reasonable efforts, including informing users not to host content spreading misinformation or impersonating others. Following the November 2023 advisory, another was issued in December 2023, urging all internet intermediaries to adhere strictly to the IT Rules, 2021, particularly emphasizing the provisions of Rule 3(1)(b). It is noteworthy that the recent AI advisory explicitly references the December advisory.

The immediate response to the latest AI advisory was met with sharp criticism from various sectors, raising concerns regarding its legal validity. Many startups expressed apprehension, highlighting its potential to disproportionately hinder their innovation capabilities, while favouring only large corporations equipped with testing resources and government approval. However, in a tweet on March 4th, IT Minister, Mr. Rajeev Chandrasekhar clarified that the advisory solely targets significant platforms, excluding startups from its purview. In a subsequent tweet on the same day, he elaborated on the potential liabilities that internet intermediaries will face under India’s criminal laws and the loss of safe harbour protections associated with unlawful content. Chandrasekhar suggested that platforms deploying AI platforms onto the public Internet could mitigate these risks by seeking prior government permission.

Most recently, however, the Government issued a new advisory on March 15, that supersedes the previous contentious advisory. The new advisory has removed the permission requirement for untested AI models.