AI undergoing testing will have to be approved before being deployed, government said - stop illegal content

Before being used, artificial intelligence (AI) that is in testing needs to get government approval. Advisories for social media and other platforms have been released by the government. Days after Google's AI tool Gemini caused a stir with its answers to inquiries concerning Prime Minister Narendra Modi, the advisory was released.

The Ministry of Electronics and Information Technology, in an advisory issued on March 1, said AI models undergoing testing should be labelled as 'unreliable'. Prevent illegal content. If this is not followed, a warning of criminal action has been given.

The advisory said that the respective platforms, intermediaries and enabling software will be held responsible for any violation. All platforms must ensure users do not use Artificial Intelligence models/Generative AI, software or algorithms to publish, transmit, store or update any unlawful content.

It is noteworthy that Google had to face criticism for the objectionable response of the AI tool Gemini on the question related to PM Modi. Union Minister Rajiv Chandrashekhar had warned that Gemini's response was a violation of IT rules as well as several provisions of the Criminal Code.

Google clarified that Gemini is undergoing testing and the company has worked quickly on this issue. On this Rajiv Chandrashekhar said, “I advise all platforms to communicate this to users and obtain their consent before deploying any platform undergoing testing on the Internet. No one can avoid accountability by apologizing later "

On Saturday, the All India Bar Association (AIBA) demanded strict punitive action against Google for allegedly spreading misinformation about PM Modi. In a representation, senior lawyer and AIBA President Adish Aggarwal claimed that Google's AI tool Gemini gave false information about the Prime Minister. Aggarwal said that AIBA has also written to the Prime Minister's Office (PMO) for action against Google.