Proposal for a Regulation of the European Parliament and of the Council laying down harmonised rules on artificial intelligence (Artificial Intelligence Act) and amending certain Union legislative acts - Analysis of the final compromise text with a view to agreement
This document is a proposal for a regulation of the European Parliament and of the Council laying down harmonized rules on artificial intelligence, commonly known as the Artificial Intelligence Act. It establishes a comprehensive legal framework for the development, placing on the market, putting into service, and use of AI systems within the Union based on a proportionate risk-based approach. The regulation prohibits certain manipulative and exploitative AI practices, sets mandatory requirements and conformity assessment obligations for high-risk AI systems, introduces transparency obligations for specific AI systems, and establishes governance structures at national and EU levels, including the European AI Office.
Summary and content fields are drawn from the document text and are pending editorial review.
Who issued it
Who it governs
AI developers, providers, deployers, importers, distributors, and regulators operating within or placing outputs into the European Union internal market need to read this to understand their legal compliance obligations.
What the document contains
The document comprehensively covers harmonized rules for AI systems and general-purpose AI models across the European Union. It addresses prohibited AI practices, classification rules and mandatory requirements for high-risk AI systems, obligations for various operators along the AI value chain, transparency obligations, innovation-support measures like regulatory sandboxes, governance structures, and enforcement mechanisms including penalties.
Military, defense, and national security applications of AI are entirely excluded from the scope of the regulation, as are AI systems and models developed exclusively for scientific research and development.
- The regulation implements a risk-based approach, categorizing AI systems into unacceptable risk (prohibited), high-risk, and minimal or specific transparency risk.
- Certain AI practices are strictly prohibited, such as manipulative techniques that distort human behavior, social scoring by public or private actors, and specific types of real-time biometric identification in public spaces by law enforcement.
- High-risk AI systems must adhere to mandatory requirements regarding risk management, data governance, technical documentation, record-keeping, transparency, human oversight, accuracy, robustness, and cybersecurity.
- Providers of general-purpose AI (GPAI) models have specific obligations, including maintaining technical documentation, respecting copyright law, and providing training content summaries, with stricter rules for models presenting systemic risks.
- A governance framework is established featuring the European AI Office, the European Artificial Intelligence Board, a scientific panel of independent experts, and an advisory forum.
· Title I: General Provisions · Title II: Prohibited Artificial Intelligence Practices · Title III: High-Risk AI Systems · Title IV: Transparency Obligations for Providers and Deployers of Certain AI Systems · Title V: Measures in Support of Innovation · Title VI: Governance · Title VII: EU Database for High-Risk AI Systems Listed in Annex III · Title VIII: Post-Market Monitoring, Information Sharing, Market Surveillance · Title VIIIa: General Purpose AI Models · Title IX: Codes of Conduct · Title X: Confidentiality and Penalties · Title XI: Delegation of Power and Committee Procedure · Title XII: Final Provisions
How it scores
Each score is backed by a quote verified against the document text. Editorial review has not yet confirmed them.
Privacy
4/5Protection of personal data and private life
The right to privacy and to protection of personal data must be guaranteed throughout the entire lifecycle of the AI system.
Recital 45a
screening score · 2026-10-23