EU AI Act Transparency Obligations Take Effect on August 2
AI 안전 | Tue Aug 04 2026 00:00:00 GMT+0000 (Coordinated Universal Time) | 1 sources
The EU AI Act's transparency provisions took effect, mandating chatbot disclosures and deepfake labeling.
Analysis
[European Union] enacted AI Act transparency rules [1]
- Mandatory disclosure of interactions with AI models starting August 2
- Requirement to indicate AI-generated or altered content
- Aimed at easier identification of chatbots and AI deepfakes
[AI Providers] required to embed machine-readable markings on synthetic content [1]
- Applies to companies developing and selling AI systems
- Must explicitly inform users when interacting with AI
- Must embed machine-readable markers in synthetic audio
- images
- video
- and text
[AI Deployers] enforced deepfake labeling obligations [1]
- Applies to platforms and services using AI systems built by providers
- Must label AI-generated images
- audio
- and video designed to appear real
- Some companies like Meta and SpaceXAI are classified as both providers and deployers
[European Commission] introduced standardized AI disclosure labels [1]
- Developed a set of AI disclosure labels for tech platforms to adopt
- Similar to existing labels on TikTok
- and Facebook
- Use of label icons is optional but labeling requirements themselves are mandatory
[AI Act Enforcement] imposed fines of up to €15M or 3% of global revenue for non-compliance [1]
- Applies immediately to new AI systems
- Models released before August 2 have a four-month grace period until December 2
- Fines up to approximately $17.2M or 3% of global annual revenue