Today, a law on artificial intelligence came into force in the European Union, which will regulate its use, UNN reports , citing a statement by the European Commission.
Today, the European Artificial Intelligence Act (AI Act), the world's first comprehensive regulation on artificial intelligence, comes into force. The AI Act is designed to ensure that artificial intelligence developed and used in the EU is trustworthy, with safeguards to protect fundamental human rights. The Regulation aims to create a harmonized internal market for artificial intelligence in the EU, encourage the adoption of this technology and create a favorable environment for innovation and investment
The EU Artificial Intelligence Law defines several categories of risks for AI systems:
- Minimal risk: systems such as recommendation and spam filters are not obligated under the Law due to their minimal risk to citizens. Companies may voluntarily adopt additional codes of conduct.
- Specific transparency risk: chatbots and AI-generated content should clearly inform users about the interaction with the machine. Synthetic content should be labeled in a machine-readable format.
- High risk: High-risk systems, such as recruitment or credit assessment, must meet strict requirements for data quality, logging, human oversight, and cybersecurity.
- Unacceptable risk: Systems that threaten fundamental rights, such as those that manipulate behavior or use real-time biometric identification, will be prohibited.
The Law also regulates general-purpose AI models used to perform a wide range of tasks, ensuring transparency and consideration of systemic risks.
Member States have until August 2, 2025, to designate national competent authorities that will oversee the application of the rules for artificial intelligence systems and carry out market surveillance activities. The European Commission's Artificial Intelligence Office will be the key body for the implementation of the Artificial Intelligence Law at the EU level, as well as the implementer of the rules for general-purpose artificial intelligence models
Companies that violate the law will be fined: up to 7% of global annual turnover for prohibited AI applications, up to 3% for other violations, and up to 1.5% for providing incorrect information.
Most of the rules will come into force on August 2, 2026. Prohibitions for AI systems with unacceptable risk will take effect in six months, and rules for general-purpose AI models will take effect in 12 months.
Recall
In February of this year, 27 EU member states unanimously approved the world's first comprehensive set of rules for artificial intelligence. This is the first attempt to regulate AI, according to European Commissioner for Internal Market and Services Thierry Breton