National and international legislation regarding AI

If you want to use or develop (generative) AI, you need to take various factors into account. One such factor comprises national and international legislation and guidelines.

National legislation and guidelines

Currently, Belgian legislation regarding AI is still quite limited. However, a Flemish Artificial Intelligence Plan was launched in 2019, based on the Flanders Artificial Intelligence Policy Plan. They focus on research, the digitisation of enterprises, and the AI literacy of Flemish citizens. In 2022, the federal government followed up with the National Convergence Plan for the Development of Artificial Intelligence. This plan sets out nine concrete objectives to make Belgium a “SmartAINation”. Neither plan imposes any obligations, but their planned actions demonstrate an ambition to use AI responsibly.  

Guidelines are also offered as part of the Flemish AI Strategy. For further information and recommendations at Flemish level, please consult the website of the Knowledge Centre Data & Society.

European legislation and guidelines

At the European level, the AI Act is of great importance in addition to the General Data Protection Regulation (GDPR) and copyright laws (EU and Belgium) . The AI Act is the first comprehensive legal framework related to AI and aims at making Europe a global leader in trustworthy AI.

More information about the impact of the AI Act on the development of AI (for research purposes) can be found here.

Would you like to explore the AI Act or find out how it applies to your research?

Then you can consult the Future of Life Institute's (FLI) website on The EU Artificial Intelligence Act. Here you will find both an AI Act Explorer and a Compliance Checker. Do not rely solely on this, but do make sure you also go through the AI Act itself. However, this can serve as a support tool to get a preliminary impression. Make sure to avoid the prohibited AI practices. 

The Knowledge Centre Data & Society has also developed a tool that supports you in making an initial analysis of the possible impact of the EU AI Act on your project. In six steps, the tool guides you in identifying the type of AI system (AIS) or model (AIM) you are working with, whether it falls within the scope of the AI Act, and what your respective role(s) and obligations may be. 

 

The AI Act is part of a broader set of policies to support the development of trustworthy AI. The following guidelines are of great importance for…

… developing (Gen)AI:

These guidelines describe fundamental rights, four ethical principles, seven key requirements, and a checklist for developing and using trustworthy AI. Read a summary of the Ethics Guidelines.

This tool translates the seven key requirements into a practical checklist with reflective questions for developers and users of AI. Read more information about the ALTAI. 

… using (Gen)AI for research purposes.

These guides provide concrete recommendations for researchers, research institutions, and funders regarding the responsible use of GenAI in research. Read a summary of the recommendations for researchers in the Living Guidelines.

This code of conduct sets out the core principles of scientific integrity and recommends transparent reporting of research methods and AI use in accordance with the standards in the field. Read more about the ALLEA code of conduct. More info about the ALLEA code of conduct and (Gen)AI? Read the Research Tip "GenAI: Translating Research Integrity into Responsible (Gen)AI Use".