The EU Artificial Intelligence Act (AI Act) is the world's first international legal act on the use of artificial intelligence. Strict regulations will come into force on February 2, after a transition period.

According to the EU AI Act, from this day onward, it is prohibited to use artificial intelligence systems that pose “unacceptable risks” to safety, health, or fundamental rights, except in cases related to “national security”.

Thus, among other things, the European Union prohibits the use of artificial intelligence software that assesses a person's social behavior. This method, known as social scoring, is used, for example, in China, where citizens are categorized based on their behavior and are rewarded or punished accordingly. The EU also does not allow the use of emotion recognition in workplaces or educational institutions.

Key restrictions:

• Manipulative artificial intelligence systems that use deceptive technologies to influence human behavior are prohibited.

• Artificial intelligence systems that exploit the vulnerability of individuals or social groups are prohibited. This category includes, for example, automated “robo-calls” that can be used to scam elderly people.

• Facial recognition in public places is also banned, including through the use of video surveillance cameras, but there are exceptions: police and other law enforcement agencies will be able to use such technologies to investigate human trafficking and terrorism crimes.

Sources:
02.02.2025., Latvijas Vēstnesis
05.02.2025. Jauns.lv