AI Act FAQ

AI Deepfakes: Frequently Asked Questions

5 questions answered with specific EU AI Act article references. 98 days until the August 2, 2026 enforcement deadline.

Not sure if your AI system is affected? Take the 5-minute diagnostic.

What is a deepfake under the AI Act?

AI-generated or manipulated image, audio, or video content that appreciably resembles existing persons, objects, places, or events and would falsely appear to be authentic (Article 3(60)).

Must deepfakes be labeled?

Yes. Deployers must disclose that content has been artificially generated or manipulated (Article 50(4)). The disclosure must be clear and visible.

Are artistic deepfakes exempt?

Partially. For 'manifestly artistic, creative, satirical, fictional analogous work,' the disclosure obligation is limited to existence of AI generation without impairing the work's display (Article 50(4)).

What about deepfake detection tools?

AI systems designed to detect deepfakes are not themselves subject to deepfake disclosure obligations. They may be classified based on their own use case.

What are the penalties for unlabeled deepfakes?

Failure to label deepfakes violates Article 50(4), which carries fines up to EUR 15 million or 3% of global turnover under Article 99(2).

Related intelligence

Get the complete AI Act FAQ as a PDF

All questions and answers in one document. Free.

No spam. Unsubscribe anytime.

Pro tier launching June 2026. Browse all briefings