Compliance Radar · AI rules
EU AI Act transparency duties for chatbots and AI-generated content apply from August 2, 2026
Chatbots and AI-generated content now need disclosures in the EU.
- Status
- In force
- Takes effect
- August 2, 2026. Existing generative systems: marking by December 2, 2026. High-risk: December 2, 2027 and August 2, 2028.
- Announced
- AI Act (2024); Digital Omnibus on AI adopted July 8, 2026, in force July 27, 2026
- Type
- law
What changed
Chatbots must say they're AI unless it's obvious, generative output must be machine-readably marked, and published deepfakes must be disclosed. Fines reach €15 million or 3% of turnover.
Who it affects
Providers and deployers of AI used in the EU, including sites with chatbots or AI media.
What to do
- Open every chatbot with a clear AI notice.
- Label deepfakes you publish.
- Ask vendors how they meet the marking duty.
What ComplyMo already handles
Page audits and daily re-checks: Growth and up. Everything else: every plan, from $19 a month.
Sources
- EUR-Lex: Regulation (EU) 2026/1744 (Digital Omnibus on AI) (opens in a new tab)
- EUR-Lex: Regulation (EU) 2024/1689 (AI Act) (opens in a new tab)
- White & Case: AI Omnibus in force (opens in a new tab)
Summary of public sources, not legal advice.