Written by Tiffany Cunha, GRC & Data Protection Specialist at Palqee Technologies
The European Union has introduced the EU Artificial Intelligence Act, a comprehensive law that regulates the use of AI systems in the EU. In this series, ‘Decoding AI: The European Union’s Take on Artificial Intelligence’, we break down everything you need to know about the law for you.
Get your free copy of Palqee’s EU AI Act Framework Get early access to the BETA program of #PAM. The AI observability solution for AI systems. |
Some background on the EU AI Act
The journey of the European Union's AI Act has been a thorough and meticulous one, starting back in April 2021 when the European Commission laid down its proposal. This proposal aimed to create a unified framework to govern the deployment, operation, and sales of AI systems within the EU.
Since then, there have been significant strides in shaping this legislation. Notably, in December 2023, a provisional agreement was reached between the Parliament and the Council, setting a solid foundation for the Act's progression.
However, the most significant advancements occurred in 2024. In February, two key developments occurred that underscored the EU's commitment to regulating artificial intelligence comprehensively and effectively.
Firstly, on February 21, 2024, the European Artificial Intelligence Office was officially launched within the European Commission's Directorate-General for Communication Networks, Content, and Technology. This new office is set to play a crucial role in overseeing the implementation of the AI Act, with a particular focus on general-purpose AI, ensuring that the regulations are applied consistently and effectively across the EU.
Secondly, on February 13, 2024, the Internal Market and Civil Liberties Committees took a decisive step by voting 71-8, with 7 abstentions, to approve the negotiated outcome with the member states on the AI Act. This vote was a clear indication of the strong support within the European Parliament for the Act's objectives and provisions.
Next Steps
With the foundational structures in place and the European Parliament's approval, the EU AI Act is poised for its next phases towards full enactment and application.
Official Enactment: The AI Act is anticipated to become official law by mid-2024. This final step will mark the transition from proposal and negotiation to actionable legislation, setting the stage for its provisions to take effect across the European Union.
Phased Implementation: Once enacted, the AI Act's provisions will be implemented in stages. The initial focus will be on prohibiting certain AI systems, with a mandate for member states to enforce these bans within six months.
Broader Application: After one year from the Act's enactment, rules for general-purpose AI systems will be applied, broadening the scope of regulation and oversight. By mid-2026, the entire spectrum of the AI Act's regulations, especially those concerning high-risk AI systems, will be in full effect, marking a significant milestone in the EU's AI governance.
AI Office's Role: The European Artificial Intelligence Office will spearhead the supervision and enforcement of the Act, particularly for general-purpose AI systems, acting as a central body to ensure uniform application of the law across the Union.
What does the EU AI Act regulate?
The EU AI Act provides us with:
Harmonised rules for the placing on the market, the putting into service and the use of artificial intelligence systems (‘AI systems’) in the Union,
Prohibitions of certain artificial intelligence practices,
Specific requirements for high-risk AI systems and obligations for operators of such systems,
Harmonised transparency rules for AI systems intended to interact with natural persons, emotion recognition systems and biometric categorisation systems, and AI systems used to generate or manipulate image, audio or video content, and
Rules on market monitoring and surveillance.
Key Features of the AI Act
The AI Act classifies AI systems into the following categories:
Prohibited AI systems: It prohibits applications and systems that create unacceptable risks, such as government-run social scoring systems,
High-risk AI systems: applications, like CV-scanning tools for job ranking, are subject to specific legal requirements, and
Applications not explicitly banned or classified as high-risk, such as the systems that interact with humans (chatbots) or generate or manipulate content (deep fakes).
This balanced approach aims to address potential risks associated with AI technology while fostering innovation.
To whom does it apply?
The Regulation applies to:
Providers who make AI systems available on the market or put them into service within the Union, regardless of whether they are based in the Union or a third country,
Users of AI systems located within the Union, and
Providers and users of AI systems located in a third country if the output generated by the system is used within the Union.
What is the impact to international companies in the EU?
To international companies this means that if they offer AI systems for sale or use them within the European Union, they are subject to the Regulation, regardless of whether they are based in the EU or in another country.
It also means that if they have users of AI systems located within the EU, they need to comply with the Regulation. Furthermore, if they are located in a third country but their AI system's output is used within the EU, they are also covered by the Regulation.
In summary, the Regulation has a broad reach and applies to both providers and users of AI systems, irrespective of their location, as long as there is an impact or connection to the European Union. International companies operating in the EU must ensure compliance with the Regulation's requirements.
Global Impact
Similar to how the European Union's General Data Protection Regulation (GDPR) became a global benchmark for data privacy, the EU AI Act is poised to set a new standard for AI regulation worldwide.
By establishing comprehensive guidelines for the ethical and responsible use of AI, the Act not only addresses the immediate needs within the EU but also offers a model that could inspire similar legislative efforts globally.
As the EU AI Act progresses towards full implementation, its influence is expected to extend beyond Europe, shaping the international discourse on AI and promoting practices that prioritise human welfare and sustainability.
Lastly, with the rapid advancements in AI technology, the timely introduction of this regulation demonstrates the EU's commitment to leading by example in balancing innovation with ethical considerations, potentially guiding the future direction of AI development and regulation around the world.
Комментарии