The Council of the EU gave the final go-ahead to the world's first-ever artificial intelligence (AI) regulation law. The new law aims to encourage the development and deployment of safe and reliable AI systems in the EU single market by both private and public actors. At the same time, it aims to ensure respect for citizens' fundamental rights and to stimulate investment and innovation in the field of artificial intelligence in Europe.
The AI Act only applies to areas within EU law and provides for exceptions, such as for systems used exclusively for military and defense purposes, as well as for research purposes.
"The adoption of the AI law is an important moment for the European Union. This landmark law, the first of its kind in the world, addresses a global technological challenge that also creates opportunities for our societies and economies. With the AI law, Europe emphasizes the importance of trust, transparency and accountability when working with new technologies, while at the same time ensuring that this fast-changing technology can flourish and stimulate European innovation," commented the Belgian Secretary of State for Digitization, Administrative Simplification , privacy protection and building regulation Mathieu Michel.
Classification of AI systems
The new law categorizes different types of artificial intelligence according to risk. AI systems that pose only limited risk will be subject to very light transparency regulations, while high-risk AI systems will be permitted but subject to a set of requirements and obligations to gain access to the EU market. Artificial intelligence systems such as cognitive behavioral manipulation and social scoring will be banned in the EU as their risk is deemed unacceptable.
The law also bans the use of AI for police use, based on profiling and systems that use biometric data to categorize people according to specific categories such as race, religion or sexual orientation.
The AI Act also addresses the use of General Purpose Artificial Intelligence (GPAI) models. GPAI models that do not pose systemic risks will be subject to some limited requirements, for example on transparency, but those with systemic risks will have to comply with stricter rules.
New management system
To ensure proper implementation, several new governing bodies are being created:
- Office for Artificial Intelligence within the Commission for the implementation of common rules across the EU;
- Scientific panel of independent experts to support implementation activities;
- An AI Board with representatives of Member States to advise and assist the Commission and Member States on the consistent and effective implementation of the AI Law;
- Stakeholder Consultative Forum to provide technical expertise to the AI Board and the Commission;
Fines for breaches of the AI Act are set as a percentage of the offending company's global annual turnover in the previous financial year or a predetermined amount, whichever is higher. SMEs and start-ups are subject to proportionate administrative fines.
The new law also provides that AI regulatory sandboxes, allowing for a controlled environment for the development, testing, and validation of innovative AI systems, must also allow testing of innovative AI systems in real-world settings.
next steps
After being signed by the presidents of the European Parliament and the Council, the legislative act will be published in the Official Journal of the EU in the coming days and will enter into force twenty days after that publication.
The new regulation will apply two years after its entry into force, with some exceptions for certain provisions.
The AI Act is a key element of EU policy to promote the development and single market entry of safe and legal AI that respects fundamental rights. The Commission presented the proposal for the law in April 2021.