EU Finalizes Landmark AI Act, Setting Global Rules for Artificial Intelligence

The European Union has formally approved the AI Act. This is the world’s first comprehensive law for artificial intelligence. The landmark legislation was finalized today.

It creates a legal framework for AI development and use. The rules are based on the potential risks of AI systems. This sets a new global standard for tech regulation.

How the New AI Act Classifies Risk and Bans Certain Uses

The law categorizes AI applications by risk level. It outright bans AI systems deemed an unacceptable risk. This includes social scoring by governments and predictive policing.

According to Reuters, the ban also covers emotion recognition tools in workplaces and schools. High-risk AI, like that used in critical infrastructure, will face strict obligations. These include rigorous testing and human oversight.

Lower-risk AI, such as chatbots, will have lighter transparency requirements. The rules aim to protect citizens’ fundamental rights. They also seek to foster innovation within clear boundaries.

The Global Impact and Enforcement Timeline of the Legislation

This law positions the EU as a de facto global tech regulator. Its “Brussels effect” often sees companies worldwide adopting EU standards. Non-compliance will lead to significant fines.

Violations can cost companies up to €35 million or 7% of global turnover. The rules will be phased in over the next few years. The bans on unacceptable AI will apply in just six months.

Companies are now assessing how to align their products. The law provides legal certainty for a rapidly evolving technology. It balances innovation with fundamental rights protection.

The EU AI Act establishes a precedent other nations are likely to follow. This comprehensive law will shape the future of artificial intelligence worldwide. Its impact on technology and society will be profound.

Info at your fingertips

What is banned under the EU AI Act?

The Act bans AI for social scoring by governments. It also prohibits predictive policing systems that profile individuals. Emotion recognition in workplaces and schools is forbidden.

When does the AI Act officially become law?

The law will be fully effective within 24 months. However, bans on unacceptable AI uses start in just six months. High-risk AI rules apply after one year.

What are the penalties for breaking the AI Act rules?

Fines can reach up to €35 million. Alternatively, companies may pay 7% of their total global annual turnover. The exact amount depends on the violation and the company’s size.

How does the Act affect AI developers outside Europe?

Any company offering AI systems in the EU market must comply. This includes developers based in the United States and Asia. The rules have a broad extraterritorial reach.