Wire · founder news, decoded · regulatory
EU gives more wiggle room on labelling AI deepfakes | Euractiv
◆ Published
20 July 2026
◆ Topic
regulatory
◆ Geography
◆ Source
◆ Verified
Fusion42 · 20 July 2026 · Fusion42 review
The EU Commission released final guidance on AI Act transparency rules, exempting some deepfake-adjacent content (like background replacement for aesthetic purposes) from mandatory labelling, and weakening broadcast labelling requirements from 'persistent' to 'where possible'. Advertisers lobbied successfully to avoid 'labelling fatigue' as companies face a 2 August compliance deadline.
◆ The Wire takeaway
If you build content creation or advertising tools in Europe, the line between 'must label' and 'can skip labelling' just shifted—background swaps for looks are now legal grey, and broadcast warnings don't need to persist. You have until 2 August to ship this logic, but enforcement varies wildly by member state, so your compliance risk is geography-dependent.
◆ Related on Wire
- European Commission clarifies AI transparency requirements20 July 2026
- What the EU's AI transparency requirements mean for users | News.az20 July 2026
- Get Ready For The EU AI Act Transparency Obligations For AI Systems16 July 2026
- Commission publishes guidelines on transparency obligations20 July 2026
- EU AI Act Enforcement Is Here: Chatbot Rules Live, High-Risk AI Delay Now Binding Law11 July 2026
- AI Content Labeling Enforcement Begins in 24 Days as EU Clears Compliance Code10 July 2026
◆ Topics
ai-act · deepfake-labelling · regulatory-compliance · content-moderation · eu-enforcement