The European Union Parliament recently (13 March 2024) passed a law called the Artificial Intelligence Act. The EU AI Act is the first of its kind globally, setting mandatory rules to reduce the risks associated with AI. Its aim is to help organizations use AI to its fullest potential while ensuring it is safe, trustworthy, and accessible to everyone. It was agreed upon after discussions with different countries in December 2023.
The law got approval from most members of Parliament. Out of 623 votes, 523 were in favor, 46 were against, and 49 decided not to vote.
The goal of this law is to protect important rights like democracy, the environment, and the law itself from AI that might be risky. At the same time, it wants to encourage new ideas and make Europe a leader in AI.
Here are some key points from the new law:
- Banned Applications: Certain uses of AI that could harm people's rights are not allowed. This includes things like using Facial Recognition Technology to collect people's pictures from the internet without their permission, or using AI to guess how someone is feeling at work or school. Other banned uses include AI that tries to predict crime based only on someone's personal information, and AI that tries to change how people behave or take advantage of their weaknesses.
- Special Rules for Police: The police can't use certain types of AI to identify people unless it's for very specific reasons, like finding a missing person or stopping a terrorist attack. Even then, they have to follow strict rules and get permission from a judge or someone in charge. Using this type of AI after an event, like looking at recorded footage, is considered risky and needs special permission.
- Rules for High-Risk Systems: AI systems that could cause big problems, like in healthcare or banking, have to follow strict rules. These rules include making sure the AI is safe, keeping records of how it's used, being open about how it works, and having people check on it to make sure it's doing what it should.
- Transparency Requirements: AI systems that are used for everyday things, like computer programs or models, need to be clear about where their information comes from and what it's used for. If the AI is very powerful and could cause big problems, there are even stricter rules to make sure it's safe and people know what's happening.
- Support for Innovation and Small Businesses: The law also wants to help small businesses and new ideas. It says that countries should set up places where companies can test out new AI ideas before selling them to the public. These places should be open to everyone, especially small businesses and start-ups, to encourage new ideas and make sure they're safe before being used by everyone.
Europe's new AI Act is crucial for individuals, businesses, and policymakers alike. These regulations not only ensure the safe and responsible use of artificial intelligence but also promote innovation while safeguarding fundamental rights and values. By staying informed about these laws and their implications, we can navigate the evolving landscape of technology with confidence, ensuring that AI works for the benefit of society as a whole.
