The recent adoption of the AI Act by the European Parliament marks a significant milestone in the regulation of artificial intelligence (AI) technologies. This comprehensive legislative framework sets out rules and standards aimed at ensuring the responsible and ethical development and use of AI systems across various sectors in Europe.
One of the key aspects of the AI Act is its focus on transparency, human oversight, and accountability. These principles are crucial in addressing the potential risks associated with AI technologies, such as bias, discrimination, and lack of accountability in decision-making processes. The Act clearly defines “AI system” as a machine-based system designed to operate with varying levels of autonomy, that may exhibit adaptiveness after deployment and that, for explicit or implicit objectives, infers, from the input it receives, how to generate outputs such as predictions, content, recommendations, or decisions that can influence physical or virtual environments. As such, the Act is aimed at AI system providers all over the European market.
The AI Act also follows a risk-based methodology of compliance in the following way:
- Unacceptable risk – certain AI practices are prohibited outright. For example, practices that manipulate human behaviour or exploit individuals’ vulnerabilities
- High risk – certain AI practices are scrutinised with risk-mitigation systems, high-quality data sets, logging of activity, detailed documentation, clear user information, human oversight, and a high level of robustness, accuracy and cybersecurity. For example, AI use in critical infrastructure
- Limited risk – providers of AI systems must inform natural persons that they are interacting with AI systems. For example, chatbots
- Minimal risk – there are no restrictions on minimal risk AI systems. For example, AI-enabled video games or spam filters
The AI Act works in tandem with the existing General Data Protection Regulation. European law on the protection of privacy and personal data will continue to apply in connection with the AI Act.
Final publishing of the Act is slated for May or June in the European Union’s official Journal. Parts of the Act will come into force 12 months after the law is passed, with the remainder coming into force 24 months later.
For a full reading of the Act, see here.
