EU AI Act Enters Force, Regulating High-Risk Systems

The European Union's landmark Artificial Intelligence Act, considered the world's first comprehensive legal framework for artificial intelligence, has entered into force this week. This legislation aims to establish clear rules and regulations for the development, deployment, and use of AI systems across the EU, categorizing them based on their potential risk to fundamental rights and safety. The act introduces a tiered approach, with the most stringent regulations applied to high-risk AI systems.

High-risk AI systems, as defined by the act, include those used in critical infrastructure, education, employment, essential services (like public access to justice and democratic processes), law enforcement, migration, and border control. For these systems, developers and deployers must adhere to strict requirements, including robust risk management systems, high-quality data sets to minimize bias, detailed documentation, transparency measures, human oversight, and a high level of cybersecurity. The act mandates that these systems undergo conformity assessments before they can be placed on the market or put into service. Failure to comply with these regulations can result in significant fines, with penalties reaching up to €35 million or 7% of a company's global annual turnover, whichever is higher.

While the AI Act is now in force, not all its provisions are immediately applicable. The act establishes a phased implementation timeline. Some obligations, particularly those related to general-purpose AI models and foundation models, will take effect after a longer period, allowing companies more time to adapt. For instance, obligations for general-purpose AI models are set to apply 12 months after the act enters into force. This phased approach is designed to provide businesses, especially smaller ones, with the necessary time to understand and implement the new compliance requirements without stifling innovation. The European Commission will oversee the implementation and enforcement of the act, working in conjunction with national authorities.

The AI Act's entry into force marks a significant moment in global AI governance. It signals the EU's commitment to fostering trustworthy AI that respects fundamental rights and democratic values. The legislation is expected to set a precedent for other jurisdictions considering similar regulatory approaches to artificial intelligence. The act also includes provisions for prohibited AI practices, such as social scoring by governments and AI systems that manipulate human behavior to circumvent free will, which are banned outright. Other AI systems, like those used for chatbots, will have transparency obligations, requiring users to be informed that they are interacting with an AI.