EU AI Act, Article 50: applies from 2 August 2026.
- AI tool providers must machine-mark synthetic images, audio, video and text. According to secondary sources, generative tools already on the market before 2 August 2026 have until 2 December 2026 to comply under the 2026 "Digital Omnibus" Check in-app.
- You, as a "deployer": if you publish a deepfake, meaning realistic AI image, audio or video resembling real people, places or events, you must disclose that it's AI-generated or manipulated.
- For clearly artistic, satirical or fictional work, you only need to disclose it in a way that doesn't spoil the work.
- AI-written text on matters of public interest must be disclosed unless a human has reviewed it and someone takes editorial responsibility.
- The Commission has worked on a voluntary Code of Practice on marking and labelling. Its final status is reported inconsistently Check in-app.
France: commercial content showing an AI-generated face or body needs "Image virtuelle" (or equivalent), and retouched bodies or faces need "Image retouchée".
Italy: AGCOM's code for relevant influencers requires labelling AI alterations, filters and deepfakes.
Canada: the Ad Standards 2025 guidelines ask for #MadeWithAI or #AIcreated on AI-altered images that show product benefits, and say virtual influencers shouldn't "review" experiences they can't have, like taste.
Platforms
- Meta (Instagram, Facebook, Threads): requires you to use its AI disclosure tool for photorealistic video or realistic-sounding audio that was digitally created or altered, and may apply penalties if you don't. The label is called "AI info."
- TikTok: requires labelling of realistic AI-generated content and has a creator AI label. It auto-labels some uploads using C2PA Content Credentials.
- YouTube: tick "Altered or synthetic content" in Studio for realistic AI content. Beauty filters, scripts and thumbnails don't need it. Repeated failure to disclose can lead to removal or suspension from the Partner Program.