Artificial intelligence act
The EU's AI act is the world's first law on artificial intelligence. It aims to ensure AI systems are safe, ethical and trustworthy.
What the AI act is
The AI act aims to ensure that artificial intelligence (AI) systems are developed and used responsibly. The rules impose obligations on providers and deployers of AI technologies and regulate the authorisation of artificial intelligence systems in the EU single market.
The law addresses risks linked to AI, such as bias, discrimination and accountability gaps, promotes innovation and encourages the uptake of AI.
As the world's first law regulating AI, the EU’s rules could set a global standard in AI regulation, just as the general data protection regulation (GDPR) has done for data privacy, promoting ethical, safe, and trustworthy artificial intelligence worldwide.
Higher risk, stricter rules
The AI act addresses the risks associated with specific uses of AI, categorising them into four levels of risk and establishing different rules accordingly.
Click or tap on the examples in the pyramid below to see the level of risk they pose and the rules applied to them.
Interactive diagram showing an inverted pyramid composed of four types of AI systems corresponding to different levels of risk.
The AI act's four risk levels and their corresponding rules are:
Minimal or no risks
Most AI systems do not pose risks. AI-powered games or spam filters can be used freely. They are not regulated or affected by the EU's AI act.
Limited risks
AI systems that present only limited risks, such as chatbots or AI systems that generate content, are subject to transparency obligations, such as informing users that their content was generated by AI so that they can make informed decisions concerning further use.
High risks
High-risk AI systems, such as those used in disease diagnoses, autonomous driving and the biometric identification of individuals involved in criminal activities or investigations, must meet strict requirements and obligations to gain access to the EU market. These include rigorous testing, transparency and human supervision.
Unacceptable risks
AI systems that pose a threat to people’s safety, rights or livelihoods are banned from use in the EU. These include cognitive behavioural manipulation, predictive policing, emotion recognition in the workplace and educational institutions, and social scoring. The use of real-time remote biometric identification systems such as facial recognition by law enforcement authorities in public spaces is also prohibited, with some limited exceptions.
General-purpose AI
General-purpose AI (GPAI) models power AI systems that are capable of performing a wide range of tasks, such as text generation and image recognition, across different applications.
GPAI models that do not pose systemic risks are subject to limited requirements, such as transparency obligations, while those with systemic risks must comply with stricter rules.
Governance architecture
To ensure proper enforcement, the AI act establishes several governing bodies:
- an AI office within the European Commission to enforce common rules across the EU
- a scientific panel of independent experts to support enforcement activities
- an AI board with member states’ representatives to advise and assist the Commission and member states on the consistent and effective application of the AI act
- an advisory forum for stakeholders to provide technical expertise to the AI board and the Commission
Penalties
The law establishes fines for violations of the rules.
Penalties are based on a percentage of the company's global turnover from the previous year or a set amount, whichever is higher. SMEs and start-ups face proportional fines.
Support for innovation
The AI act's objectives are not only to enhance effective enforcement of existing law on fundamental rights and safety. It also aims to promote investment and innovation in AI within the EU, and to facilitate the development of a single market for AI applications.
Accordingly, the rules include further provisions to support AI innovation in the EU. This also goes hand in hand with other initiatives, including the EU's coordinated plan on artificial intelligence which aims to accelerate investment in AI in Europe.
The work of the Council on AI
In October 2020, the European Council discussed the digital transition.
In relation to AI, EU leaders invited the Commission to:
- propose ways to increase European and national public and private investments in artificial intelligence research, innovation and deployment
- ensure better coordination and more networks and synergies between European research centres based on excellence
- provide a clear, objective definition of high-risk artificial intelligence systems
In April 2021, the Commission released a proposal for a regulation aiming to harmonise rules on artificial intelligence (AI act) and a coordinated plan which includes a set of joint actions for the Commission and member states. This package of rules aims to improve trust in artificial intelligence and foster the development and updating of AI technology.
The Council discussed the proposed act in October 2021, highlighting on the one hand the considerable societal and economic benefits that AI can bring across a wide range of sectors and on the other the need to protect privacy and ensure safety and security.
The Council adopted its position on new rules for AI on 6 December 2022. The Council wants a safe, lawful and trustworthy AI that respects fundamental rights.
On 9 December 2023, following three-day 'marathon' talks, negotiators in the Council and the European Parliament reached a provisional agreement on the AI act.
The act was formally adopted by the Council on 21 May 2024 and entered into force on 1 August 2024.
- Artificial intelligence (AI) act: Council gives final green light to the first worldwide rules on AI (press release, 21 May 2024)
- Artificial intelligence act: Council and Parliament strike a deal on the first rules for AI in the world (press release, 9 December 2023)
On 29 June 2026, the Council gave its final green light to a new regulation that streamlines and simplifies rules on artificial intelligence. The new law is part of the ‘Omnibus VII’ legislative package in the EU’s simplification agenda. A key provision of the new law prohibits AI practices involving the generation of non-consensual sexual and intimate content or child sexual abuse material.
Your life online: how is the EU making it easier and safer for you?
The European Union is actively working on improving the digital environment for the benefit of all Europeans. Our digital life needs to be safe, easy and respectful of basic freedoms.
Read our feature story to discover how the EU is creating a common eID, wants rules for ensuring a safe and trustworthy AI and is facilitating health data sharing in the European health data space.
See also
AI explained: uses and impact
Benefits and risks of AI
Digital transition: shaping the EU's digital future
Last review: 29 April 2025