India has imposed curbs on artificial intelligence content on social media platforms. The restrictions will take effect on February 20, the federal government said on Tuesday.
The curbs stipulate that AI-generated content should be prominently labelled, and also cuts the time allowed for illegal content – including non-consensual deepfakes – to be taken down, reducing this to within two-to-three hours from the earlier 36-hour window for social media platforms.
Content ruled as illegal by a court or an “appropriate government” has to be taken down within three hours. Sensitive content featuring non-consensual nudity and deepfakes must be removed within two hours, according to the new rules, The Hindu newspaper reported.
The draft version of the new rules specifies that 10% of any imagery would have to be covered in the labelled disclosure, but platforms have been given some leeway, The Hindu quoted an unnamed official as saying.
The labelling must comprise permanent markings that can’t be removed.
Indian IT law states that failure to comply with the rules will result in the loss of safe harbor protection – the principle that sites that allow users to post content are not held liable for them.
The development also comes ahead of the India AI Impact summit, which will be held in New Delhi next week.
India says its emphasis on AI governance balances innovation with accountability.
India is the world’s third-largest AI talent hub and leads globally in AI skill penetration, according to Stanford’s AI Index.
Google, Amazon, and Microsoft together committed more than $50 billion in investments in India toward cloud and AI infrastructure last year. New Delhi has extended a tax holiday in its federal budget for companies building data center infrastructure in the country to serve global markets.