All content created using AI must be labelled, deepfake photos and videos must be removed within 3 hours; New rules come into effect today
If AI is used to promote child pornography, obscenity, fraud, weapons-related information, or impersonation, it will be considered a serious crime.
There was an uproar over the deepfake video case of actress Rashmika Mandanna way back in November 2023. An FIR was lodged under Sections 465 (forgery) and 469 (forgery with intent to harm reputation) of the Indian Penal Code. Additionally, Sections 66C and 66E of the Information Technology Act were invoked.
New Delhi: If any photo, video, or audio has been created using AI, it has been made mandatory to label it. Social media platforms will also be required to remove any objectionable content within 3 hours of receiving a complaint. These new rules come into effect on February 20, 2026. A notification was issued on February 10.
Post will be deleted if metadata is tampered with
AI Label; A 'Digital Stamp' on Videos: Just as food packets indicate whether it's 'vegetarian' or 'non-vegetarian,' every AI video, photo, or audio will now have a label. Suppose you created a video using AI in which a leader is giving a speech. It should clearly state, "AI Generated," in the corner of the video.
Technical Markers; Digital DNA: Metadata can be considered the file's 'digital DNA.' It's not visible on the screen, but hidden within the file's coding. It will contain information about the date the photo or video was created, the AI tool it was created with, and the platform it was first uploaded on. If someone commits a crime using AI, police will be able to trace its source through this 'technical marker.'
Tamper Prohibition; Labels Cannot Be Erased: Previously, people would remove the watermark from AI-generated photos by cropping or editing the corners to make them appear real. Now, the government has made this illegal. Social media platforms must adopt technology that will delete the content if someone tries to remove the label or metadata.
Strict action against child pornography and deepfakes
If AI is used to promote child pornography, obscenity, fraud, weapons-related information, or impersonation, it will be considered a serious crime.
3-hour deadline
Following the new IT regulations, social media companies will now have significantly less time to take action. The previous 36-hour window to remove illegal content has been reduced to just 3 hours.
Now, whenever a user uploads something on social media, the platform will be required to obtain a declaration confirming whether the content was created by AI. Companies will be required to deploy tools that can verify this claim. If a platform allows AI content to be published without disclosure, it will be held responsible.
For breaking news and live news updates, like us on Facebook or follow us on Twitter and Instagram. Read more on Latest India News on India.com.