Synthetic media laws tighten platform and tool obligations

The US TAKE IT DOWN Act criminalises publishing nonconsensual intimate imagery, including AI-generated digital forgeries. Platform obligations took effect on 19 May 2026, requiring covered services to provide a notice-and-removal process and remove reported content within 48 hours of a valid request. The Federal Trade Commission oversees platform compliance, while the Department of Justice handles criminal cases. Platforms need reliable reporting channels and ways to locate copies.

The European Union’s AI Act transparency obligations apply from 2 August 2026. Deepfakes must be disclosed as artificially generated or manipulated, and generative system providers must mark outputs in a machine-readable format. Systems already on the market have until 2 December 2026 to meet the marking requirement, but the disclosure duty was not extended. Missing labels should be treated as unknown, not proof of authenticity.

In the United Kingdom, sharing or threatening to share nonconsensual intimate images, including deepfakes, is a criminal offence. These became priority offences in September 2024, obliging platforms to prevent and remove them proactively. A further offence targeting creation of sexually explicit deepfakes, carrying an unlimited fine, was introduced through the Data (Use and Access) Act 2025.

Tennessee’s ELVIS Act took effect on 1 July 2024, extending publicity rights to a person’s voice and making unauthorised AI simulations of voice or likeness actionable. Across these regimes, labelling duties fall on generators and removal duties on hosts, making detection and timely response central compliance capabilities.