European Union Artificial Intelligence Act

PP
Pointon Partners

Contributor

Pointon Partners is a medium-sized legal firm known for its full-service offerings to businesses and stakeholders. With a focus on building long-term relationships, the firm helps clients achieve successful outcomes. They provide top-tier expertise with a personalized touch, serving a wide range of clients from Australian companies to private individuals. Additionally, they are a member of LAWORLD, offering international legal support.
Act has broad extra-territorial scope and applies to AI providers in the EU or in another country.
Australia Technology
To print this article, all you need is to be registered or login on Mondaq.com.

In this digital age, it is interesting to note that legislation known as the “Artificial Intelligence Act” (“Act”) was recently approved by the European Parliament on March 13, 2024.1

The Act aims “to promote the uptake of human centric and trustworthy artificial intelligence (AI) while ensuring a high level of protection of health, safety, fundamental rights … including democracy, rule of law and environmental protection, to protect against the harmful effects of AI systems in the Union, and to support innovation”.2  The legislation has broad extra-territorial scope and applies to AI providers in “a non-discriminatory manner”, irrespective of whether they are established in the Union or in a third country. 3

Broadly speaking, the Act identifies three categories of AI systems and establishes obligations based on their risks and potential impact.4 The three categories are:

  1. General purpose AI systems which refer to “foundation models” that display significant generality and can perform a diverse range of tasks.5 An example is ChatGPT.6
  2. High risk level AI systems which refer to the use of AI technology in specific areas such as remote biometric identification systems, critical infrastructures, education, employment, essential services, law enforcement, migration, justice and democratic processes.7 High risk AI applications also include those that are used in products falling within the scope of the EU's product safety legislation, such as machinery, toys, medical devices and lifts.8
  3. Unacceptable risk level AI systems which are expressly prohibited under the Act, subject to limited exceptions. Examples of prohibited practices include the use of AI for purposes such as emotion recognition in workplace and school contexts, social scoring, behavioural manipulation and untargeted scraping of facial images (e.g. from the internet or CCTV footage).9

Violations of the Act trigger significant penalties, with fines up to the higher of 35 million EUR or 7% of annual turnover. 10 By way of comparison, the maximum penalty for breaches of the General Data Protection Regulation (EU) is 20 million EUR or 4% of the annual revenue.11

Upon receiving formal endorsement by the European Council, the law will be published in the Office Journal of the European Union and come into force twenty days after publication.12

Footnotes

1 European Parliament, ‘Artificial Intelligence Act: MEPs Adopt Landmark Law' (Press Release, 13 March 2024). https://www.europarl.europa.eu/news/en/press-room/20240308IPR19015/artificial-intelligence-act-meps-adopt-landmark-law.

2 European Parliament, ‘Corrigendum on the position of the European Parliament and of the Council laying down harmonised rules on artificial intelligence (Artificial Intelligence Act) (Parliament Document, 16 April 2024) https://www.europarl.europa.eu/doceo/document/TA-9-2024-0138-FNL-COR01_EN.pdf, 3. 

3 Ibid 21.

4 Ibid 26.

5 Council of the European Union, ‘Proposal for a Regulation of the European Parliament and of the Council laying down harmonised rules on artificial intelligence (Artificial Intelligence Act)' (COM(2023) 812 final) (Council Document ST-5662-2024-INIT, 26 January 2024) https://data.consilium.europa.eu/doc/document/ST-5662-2024-INIT/en/pdf, 58.

6 European Parliamentary Research Service, ‘The European Union and International Law: An Overview' (Briefing, March 2023) https://www.europarl.europa.eu/RegData/etudes/ATAG/2023/745708/EPRS_ATA(2023)745708_EN.pdf, 1.

7 European Parliament, ‘Corrigendum on the position of the European Parliament and of the Council laying down harmonised rules on artificial intelligence (Artificial Intelligence Act) (Parliament Document, 16 April 2024) https://www.europarl.europa.eu/doceo/document/TA-9-2024-0138-FNL-COR01_EN.pdf, Annex III, 423-429.

8 Ibid 46.

9 Ibid 180-182.

10 Ibid 390.

11 General Data Protection Regulation, Regulation (EU) 2016/679, Art 83.

12European Parliament, ‘Artificial Intelligence Act: MEPs Adopt Landmark Law' (Press Release, 13 March 2024). https://www.europarl.europa.eu/news/en/press-room/20240308IPR19015/artificial-intelligence-act-meps-adopt-landmark-law.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.

See More Popular Content From

Mondaq uses cookies on this website. By using our website you agree to our use of cookies as set out in our Privacy Policy.

Learn More