A three-hour deadline can reshape the internet. Under India’s new AI r ules, platforms must remove flagged content almost immediately or risk losing legal protection. The goal is to combat deepfakes and synthetic generative information (SGIs). The risk is creating a system where deletion becomes the safest response, even when speech is lawful. The concern driving these rules is real. Generative AI has made it easier to manipulate images, clone voices and fabricate speeches at scale.

Impersonation, fraud, and non-consensual imagery demand firm enforcement. Clear labelling of AI-generated content can also serve as a reasonable transparency measure. But regulation is defined not just by its intent, but by its incentives. The new framework reduces the takedown window from thirty-six hours to just three hours. When safe harbour protections hinge on speed, platforms will prioritise compliance over deliberation. Remove first, assess later. In heated political climates, complaints can b e weaponised. Satire or criticism may disappear not due to illegality, but due to liability. [1] [2] That incentive structure matters. An internet governed by accelerated liability risks is becoming an........

© The Statesman