The European Union (EU) is launching a “crash test” system for artificial intelligence. The goal is to ensure that new developments with this technology are safe before they hit the market.
The bloc announced on Tuesday the creation of four permanent experimental and test laboratories throughout Europe. The initiative involved the investment of 220 million euros (about 240 million dollars), reported Bloomberg.
The test centers, which will be both virtual and physical, They will open their doors next year, but some services will already be available from next July. The expectation is that technology companies can test their artificial intelligence and robotics systems related to four areas: agri-food, health, manufacturing and smart cities.
Lucilla Sioli, director of artificial intelligence and digital industry at the European Commission, stressed at an event in Copenhagen that they also hope to address some risks associated with these developments, such as disinformation. The goal is to help developers create “reliable products more efficiently and facilitate their adoption in Europe,” the European Commission added in a statement. release.
Test centers to boost EU artificial intelligence law
The test centers are a version of the European testing system for new cars, the Technical University of Denmark said. The institution, which will lead one of the centers, explained in a press release that they will act as a “security filter» between providers and users. They also hope to influence the development of related public policies.
“AI-powered solution providers have the opportunity to test their products in real environments to assess whether they meet customer needs,” said Valentina Ivanova, AI-Matters project coordinator and initiative participant. “By offering access to testing and experimentation infrastructure across Europe, we aim to accelerate the adoption of these solutions in the market.”
The idea, as explained by the European Commission, is that the testing centers also collaborate with the implementation of the EU Artificial Intelligence Law. The European Parliament approved this month by a large majority a final draft of the regulation project for this technology in this region. Negotiations continue with the European Commission and member countries to give validity and legal validity to the regulations before 2026.
The law proposes the creation of a categorization system for artificial intelligence, based on the potential risk they pose to citizens. It also proposes a ban on general remote biometric identification systems and biometric classification.