EU Member States Approve Landmark AI Regulation
EU member states have given their approval to a proposed law that aims to regulate artificial intelligence (AI) models. The legislation, known as the AI Act, was initially introduced by the European Commission in 2021. After reaching a provisional agreement on the law’s content last December, it has now received the greenlight from EU member states, paving the way for a vote in the European Parliament.
The AI Act seeks to ban harmful AI use cases, including emotion recognition in workplaces and educational institutions. It also includes rules to regulate the use of high-risk neural networks in sectors such as insurance and banking. Additionally, the legislation introduces requirements for developers of high-impact general AI models, known as GPAI. These requirements include checking for systemic risks and mitigating any issues, as well as reporting the energy efficiency of neural networks.
Following the provisional approval of the AI Act in December, a negotiation process took place involving several EU member states. Germany and France expressed concerns about the law’s impact on AI innovation, while Austria had reservations about its privacy provisions. In response, the European Commission introduced pro-innovation measures and established the Artificial Intelligence Office as a regulatory body responsible for implementing the AI Act.
An expert group comprising officials from EU member states will advise the European Commission on the implementation process. Their priorities will include ensuring that there’s no overlap between the AI Act and other regulations, particularly those related to medical devices and machinery.
The AI Act will now proceed to a committee vote in the European Parliament later this month with expectations for a successful passage. Most officials involved in drafting the legislation believe that it won’t undergo any changes from its current version.
This regulatory development is significant for the EU as it sets a precedent for controlling and addressing the risks associated with AI. By establishing clear guidelines and restrictions, the AI Act aims to strike a balance between promoting innovation and protecting individuals’ rights and privacy.
The approval of the AI Act by EU member states marks a crucial step towards comprehensive regulation of AI in the region. As AI continues to advance and permeate various industries, governments worldwide are recognizing the importance of creating legal frameworks to ensure responsible and ethical AI development and deployment. The EU’s efforts in this regard set an example for other countries to follow and signify the growing recognition of AI’s impact on society.
