Major social media platforms, including Meta, TikTok, X, Reddit, and Snapchat, are now legally required to remove deepfake pornography and non-consensual intimate imagery within 48 hours after receiving a valid report from victims.
The requirement comes under the newly enforced Take It Down Act, a federal law designed to combat the rapid spread of AI-generated sexual deepfakes and revenge porn online. The legislation officially entered full enforcement this week after platforms were given a one-year compliance window to build reporting and removal systems.
Under the law, online platforms that fail to remove reported content within the mandated timeframe can face fines of up to $53,088 per violation. The rules apply to both authentic intimate images shared without consent and AI-generated sexually explicit deepfakes.
The Federal Trade Commission is responsible for enforcing the law and has already issued compliance reminders to major technology companies ahead of the enforcement deadline. Companies reportedly contacted by regulators include Meta, TikTok, X, Amazon, Microsoft, Reddit, Snap, and Alphabet.
The law represents one of the strongest federal responses yet to the growing abuse of generative AI tools used to create fake explicit images and videos. Deepfake technology has become increasingly accessible over the past two years, allowing users to generate realistic non-consensual content with minimal technical expertise.
Victims of deepfake abuse have long struggled to force platforms to remove harmful content quickly, especially when images spread across multiple services simultaneously. The new law requires companies to provide clear reporting tools and act rapidly once valid takedown requests are submitted.
The legislation was partially inspired by several high-profile deepfake abuse incidents involving celebrities, students, and minors. In recent years, AI-generated explicit images targeting public figures and private individuals have spread widely across social media platforms, sparking pressure for stronger regulation.
Researchers continue warning that deepfake abuse is evolving faster than current detection systems. Several recent academic studies found that many AI-generated manipulations remain difficult for both humans and automated tools to identify consistently, particularly across video and audio formats.
Governments in both the United States and Europe are now increasing pressure on technology companies to respond more aggressively to AI-generated abuse, especially content involving harassment, blackmail, sexual exploitation, and non-consensual imagery.