Artificial Intelligence law at Romania
As of May 2025, Romania is actively developing its legal and strategic framework for Artificial Intelligence (AI), aligning with European Union (EU) initiatives. Here's an overview of the current developments:
🇪🇺 Alignment with the EU AI Act
Romania, as an EU member state, is subject to the EU Artificial Intelligence Act (EU AI Act), which was published in the EU Official Journal on 12 July 2024 and is set to enter into force on 1 August 2024. The Act adopts a risk-based approach, categorizing AI systems into four levels: unacceptable, high, medium, and low risk. Prohibited AI practices include subliminal techniques, social scoring, and certain uses of biometric identification and emotion recognition .(Wolf Theiss - Leading Lawyers in CEE&SEE, Global Practice Guides, Leroy Law)
Key provisions of the EU AI Act include:
Transparency Obligations: Providers must inform users when interacting with AI systems and disclose synthetic content generated by AI .(Leroy Law)
General-Purpose AI Models: Developers must provide documentation and transparency, especially for models posing systemic risks .(Leroy Law)
Regulatory Sandboxes: Member states are required to establish AI regulatory sandboxes by August 2026 to foster innovation in a controlled environment .(Leroy Law)
Penalties: Violations can result in fines up to €35 million or 7% of global annual turnover .(Leroy Law)
The EU AI Act's provisions on prohibited AI practices entered into force on 2 February 2025 .(CEE Legal Matters)
🇷🇴 Romania's National AI Strategy (2024–2027)
In July 2024, Romania's government approved the National Artificial Intelligence Strategy for 2024–2027, aiming to integrate AI technologies across various sectors. The strategy focuses on:(Digital Watch Observatory)
Digital Public Administration: Enhancing AI adoption in public services, particularly by the tax authority (ANAF), for risk analysis and optimization of public spending .(Digital Watch Observatory)
Digital Economy: Promoting AI-driven economic growth and competitiveness.
Digital Education: Fostering AI literacy and education to build a skilled workforce.
Cybersecurity: Ensuring secure AI systems and data protection.
Emerging Technologies: Integrating AI with other technologies like 5G, IoT, and blockchain .(Digital Watch Observatory)
The implementation will be overseen by the Romanian Intelligence Service (SRI) and the Foreign Intelligence Service (SIE) through an inter-ministerial commission .(CAIDP Europe)
🧠 AI Literacy and Public Awareness
The EU AI Act emphasizes the importance of AI literacy, requiring providers and deployers of AI systems to ensure that their personnel and users possess sufficient knowledge to operate and understand AI systems responsibly .(CEE Legal Matters)
🔍 Proposed Legislation: Deepfake Regulation (PL-X No. 471/2023)
Romania is considering draft legislation known as the Deepfake Regulation (PL-X No. 471/2023), which aims to address the responsible use of AI in combating the deepfake phenomenon. The legislation targets the dissemination of visual and/or audio content generated or altered using technology in the context of deepfakes, with the objective of preventing misinformation and preserving the authenticity of transmitted messages. Proposed penalties for non-compliance range from €2,000 to €40,000 .(Global Practice Guides)
✅ Summary
EU AI Act Compliance: Romania is aligning its national policies with the EU AI Act, which sets a comprehensive regulatory framework for AI systems.(Wolf Theiss - Leading Lawyers in CEE&SEE)
National AI Strategy: The 2024–2027 strategy focuses on integrating AI across public administration, economy, education, cybersecurity, and emerging technologies.(Digital Watch Observatory)
AI Literacy: Efforts are underway to enhance AI literacy among the public and professionals.
Legislative Developments: Proposed legislation like the Deepfake Regulation aims to address specific AI-related challenges.
RELATED Blog
0 comments