The European Parliament has passed a landmark law on artificial intelligence, which introduces rules for AI products with varying degrees of risk in a single market. The law, approved by a majority vote, will be implemented in stages over the next two years.
The main objective of the law is to protect fundamental rights and stimulate innovation, while solidifying Europe’s leadership in the field of artificial intelligence. It includes the categorization of AI systems based on potential risks, with strict control measures for high-risk systems.
High-risk AI systems are now required to maintain a journal of their actions to ensure transparency and traceability of outcomes. This measure aims to safeguard interests in cases where AI decisions may be contested, such as in employment-related decisions.
The law has sparked debates among EU members, particularly in France and Germany, who have expressed concerns that the new regulations could impede national companies’ interests and bolster competitive advantages for the United States and China.
One key aspect of the law is the protection of individuals’ fundamental rights, which includes prohibiting the creation of facial recognition databases without specific purposes and restricting the use of emotion recognition technologies in schools and workplaces.
Additionally, the law governs the use of real-time recognition systems by law enforcement agencies, permitting it only in narrowly defined cases such as finding missing persons or preventing terrorist activities.
One year after its publication, the law will come into effect in the EU official journal concerning general AI and Management, obligating developers to adhere to transparency requirements and disclose comprehensive summaries of their content usage.
The co-authors of the law are optimistic that its adoption will mark a significant stride in mitigating risks, fostering new possibilities, combating discrimination, and enhancing transparency in the realm of artificial intelligence.
This law heralds a new era in artificial intelligence regulation, placing human values and European principles at the core of technology advancement.