EU adopts law on artificial intelligence

The European Union's new AI law, which was finally adopted on May 21, 2024, introduces stricter rules for the use of artificial intelligence (AI) in the EU. The law, or rather the EU regulation, applies directly in all member states, but is not mandatory until spring 2026. The new regulation, which was approved by the EU Parliament back in March, is intended to make the use of artificial intelligence in the European Union safer.

Even if the aim of the new regulations is primarily to protect individuals against secret surveillance, automated data collection or unrecognized AI products, it can be assumed that companies and employers will have to scrutinize some digital processes in their own operational procedures and, if necessary, correct them in order to avoid the risk of accusations or even fines.

Here are some important contents of the AI Regulation that will be relevant for companies from a legal perspective:

➡ Transparency and traceability: AI systems must be transparent so that it is possible to understand how they work. Developers must label texts, sounds and images generated by AI so as not to mislead people. This also applies to the areas of marketing, sales and recruiting.

➡ Ban on certain applications: Certain AI applications that violate EU values are now banned altogether. This includes, for example, the evaluation of social behavior ("social scoring").

➡ Facial recognition in public spaces: Facial recognition in public spaces is generally not permitted. However, exceptions apply if there is a court order, e.g. for the prosecution of criminal offenses.

➡ Environmental friendliness and safety: AI systems must be environmentally friendly and safe. High-risk applications in critical infrastructures or in the education and healthcare sectors must meet strict requirements.

Feel free to contact us if you have any questions about the content or the necessary implementation steps of the Artificial Intelligence Ordinance.