EU approves rules to regulate artificial intelligence technologies

Christian Lue/Unsplash

After months of debate, EU authorities have agreed on the Artificial Intelligence Act. It is the world's first comprehensive rules to regulate AI systems.

Here's What We Know

The document includes a ban on the use of AI to recognise people by race, gender and sexual orientation. It also introduces new transparency requirements for powerful models like GPT-4.

As noted in the European Commission, the law is designed to guarantee the safe development of AI and the protection of the rights of EU citizens. Fines of up to 7 per cent of global revenue are envisaged for violators of the rules.

Bans on the use of certain AI models will come into force in six months, transparency requirements - in 12 months, and the full set of rules - in two years.

Lawmakers also included measures to make it easier for copyright owners to protect themselves from generative AI. In addition, general-purpose AI systems should become more transparent about energy use.

Source: Wired