The deadline for taking down unlawful content after receiving a valid, reasoned order from a court or the appropriate government has been slashed from 36 hours to just three hours. The broader grievance redressal timeline, within which platforms must respond to user complaints, has also been tightened, cut from 72 hours to 36 hours. The rules go further on the question of AI-generated content itself. Platforms have also been told to take active measures against unlawful AI-generated content, a category that spans deepfakes, child sexual abuse material, non-consensual intimate images, and AI-based impersonation of real people. Beyond removal, intermediaries must inform users about the "legal consequences" of creating or circulating such unlawful synthetic content, effectively building a deterrent into the platform experience itself.