Seek approval for deploying under-trial AI models: Centre to AI platforms

New Delhi, Mar 2 (PTI) Days after Google's AI platform's response to queries on Prime Minister Narendra Modi generated controversy, the government has issued an advisory for social media and other platforms to label under-trial AI models and prevent hosting unlawful content.

In an advisory issued to intermediaries/platforms on March 1, the Ministry of Electronics and Information Technology warned of criminal action in case of non-compliance.

"All intermediaries or platforms to ensure that use of Artificial Intelligence model(s) /LLM/Generative AI, software(s) or algorithm(s) on or through its computer resource does not permit its users to host, display, upload, modify, publish, transmit, store, update or share any unlawful content," the advisory said.

All platforms, intermediaries and enabling software shall be held accountable for any breaches, it added.

"Non-compliance with provisions would result in penal consequences." The advisory comes within days of Google's AI platform Gemini making uncharitable comments about Prime Minister Modi's policies. The response of Gemini to queries about the Prime Minister led to a strong reaction from the government, with Minister of State for IT Rajeev Chandrasekhar calling it a violation of IT laws.

Union Minister for Electronics and IT Rajeev Chandrashekhar on Saturday said, "The episode of Google Gemini is very embarrassing, but saying that the platform was under trial and unreliable is certainly not an excuse to escape from prosecution".

"I advise all platforms to openly disclose to the consumers and seek consent from them before deploying any under-trial, erroneous platforms on the Indian public internet. Nobody can escape accountability by apologising later. Every platform on the Indian internet is obliged to be safe and trusted," he further said.

The advisory asked the entities to seek approval from the government for deploying under trial or unreliable artificial intelligence (AI) models and deploy them only after labelling them of "possible and inherent fallibility or unreliability of the output generated".

"Use of under-testing / unreliable Artificial Intelligence model(s) /LLM/Generative AI, software(s) or algorithm(s) and its availability to the users on Indian Internet must be done so with the explicit permission of the Government of India and be deployed only after appropriately labelling the possible and inherent fallibility or unreliability of the output generated," it said.

It further said that the " 'consent popup' mechanism may be used to explicitly inform the users about the possible and inherent fallibility or unreliability of the output generated".

The Union minister said that the government is not trying to take control in its own hands, but simply allowing any and every platform on the internet is not healthy and sustainable for the ecosystem.

The advisory is in continuation to the advisory issued on December 26, 2023, which dealt with deepfakes and misinformation. PTI