In force since 02.02.2025
AI ActHigh impact In force
AI Literacy (Article 4 AI Act)
Technical and business teams working with AI must have documented training covering model limitations, risks, and human oversight.
Back to AI Act information
Check your company's AI Act readiness
Answer a short series of questions and we will prepare a recommendation and an action report.
Check your company's AI Act readiness
Answer a short series of questions and we will prepare a recommendation and an action report.
Which type of organization do you represent?
This affects the weight of obligations and implementation priorities.
The questionnaire result is indicative and does not constitute a full legal compliance assessment. Final classification requires expert legal, compliance, and technical review.
What you must do
Key obligations in chronological order — from those already in force to upcoming deadlines.
In force since 02.02.2025
AI ActCritical for selected segment In force
Prohibited AI practices (Article 5)
Ban on social scoring, manipulative AI, mass facial recognition, and emotion recognition in workplaces and schools. Continuous monitoring required.
In force since 02.08.2025
AI ActHigh impact In force
GPAI obligations — general-purpose AI models (Art. 51–56)
Providers of general-purpose AI models (chatbots, content generators, recommendation engines) must ensure transparency, training data documentation, and copyright compliance.
In force since 02.08.2025
AI ActCritical for selected segment In force
Penalty regime (Article 99)
Three-tier penalty system proportional to the type of violation:
EUR 35M
7% turnover
Prohibited practices (Art. 5)
EUR 15M
3% turnover
Other AI Act violations
EUR 7.5M
1% turnover
Supplying incorrect information
Enters into force in 112 days
AI ActHigh impact To implement
AI transparency (Article 50)
Obligation to inform users of AI interaction, mark AI-generated content (watermarking), disclose deepfakes, and notify about emotion recognition and biometric use.
Enters into force in 112 days
AI ActMedium impact To implement
High-risk AI systems (Art. 6–15, Annex III)
Full obligations for high-risk AI: risk management, data governance, technical documentation, transparency, human oversight, accuracy, and cybersecurity.
Ongoing obligation
GDPRCritical for selected segment Ongoing (before go-live and operationally)
GDPR: legal basis, notices, user rights, retention
GDPR obligations are continuous: lawful basis, information duties, rights handling, and consistent retention policies.
Open GDPR Scanner Ongoing obligation
WCAGHigh impact Ongoing (before go-live and operationally)
WCAG 2.1 AA: interface and content accessibility
Websites should maintain accessibility controls including contrast, keyboard navigation, semantic structure, ARIA, and accessible forms.
Open WCAG Scanner Enters into force in 112 days
AI ActMedium impact To implement
Human oversight for AI (Article 14)
High-risk AI systems require decision logging, human review mechanisms, incident response procedures, and the ability for human intervention.