Duration
24h Th
Number of credits
Lecturer
Language(s) of instruction
English language
Organisation and examination
Teaching in the first semester, review in January
Schedule
Units courses prerequisite and corequisite
Prerequisite or corequisite units are presented within each program
Learning unit contents
This seminar, taught in English, pursues two main objectives. First, it will outline and comment the main issues (political, social, economic and legal) related to big data and new technologies, in particular Artificial Intelligence (AI), in the European Union's legal order. Second, it will critically assess the main regulatory approaches adopted by the EU legislator, on points affecting data processing and data protection, as well as the standards imposed on designers and users of data powered technologies (such as AI).
In achieving these objectives, the students will acquire in-depth knowledge of the issues raised by the Big Data phenomenon (and the technologies having emerged as a result), of the advantages and shortcomings of the regulatory solutions given, and gain valuable insights into the future application of the EU's regulatory framework relating to AI.
The seminar will comprise 24 hours of ex cathedra teaching, structured as follows:
General information - The instructor will present the main objectives of the course and the related administrative requirements, including the examination arrangements.
Introduction - The teaching provided in this chapter addresses a simple question: why and how should new technologies be regulated? To answer this question, the teaching will proceed in three stages: (1) a brief presentation of the risks associated with artificial intelligence; (2) a brief overview of the industrial revolutions that led to the emergence of intelligent technologies; and (3) a brief presentation of the characteristics of an AI system.
Part 1 - Substantive European Artificial Intelligence Law
Chapter 1: Regulation - This chapter examines the concept of regulation sensu lato. It provides a definition of regulation in general and examines its practical modalities.
Chapter 2: The model for European AI regulation: the GDPR and its normative "progeny" - This chapter introduces and explains the impact of the GDPR on EU legislative instruments relating to new technologies, including artificial intelligence. To this end, the teaching will address four main issues: (1) the GDPR's objective of striking a balance between the free movement of data and data protection; (2) the structure of the GDPR; (3) the impact of the GDPR on subsequent legislation concerning new technologies; and (4) the emergence of European AI regulation and the extent to which its objectives and structure draw inspiration from the GDPR.
Chapter 3: The origins of European artificial intelligence regulation - This chapter examines the legislative developments relating to artificial intelligence that preceded the adoption of the AI Act. It thus provides an analysis of the ethical work that preceded its adoption and seeks to identify the policy objectives pursued by the European Commission in this context.
Chapter 4: Regulation (EU) 2024/1689 on artificial intelligence - This chapter analyses the regulatory choices made by the European legislator in the field of AI regulation. To this end, it examines the material scope of the AI Act and the risk-based approach underpinning it. A first distinction is drawn between prohibited practices, high-risk AI systems, AI systems subject to specific transparency obligations, and AI systems presenting minimal risk. A second distinction is then drawn between general-purpose AI models and those presenting systemic risk. This chapter also introduces the concepts of standardisation, certification, and market surveillance.
Part 2 - Artificial Intelligence Liability Law
Chapter 5: Selecting the Appropriate Liability Model in the Field of AI - The lectures in this chapter will pursue three objectives: (1) to critically examine the contemporary challenges that AI raises regarding the proof of harm and causation, as well as the fair allocation of risks arising from the use of AI systems; (2) to provide historical context on the ways in which traditional liability law and doctrines have responded to previous industrial revolutions, particularly industrialization and automation; and (3) drawing on these historical examples, to provide a structured and critical analysis of the expert and regulatory debates at the EU level concerning the most appropriate liability model for AI technologies.
Chapter 6: Operationalizing AI liability in EU Law - Against the backdrop of the developments discussed in Chapter 3, the lectures in this chapter will provide a critical and detailed examination of the EU law provisions that have operationalized these discussions. The two seminal legislative instruments in this context are the AI Liability Directive (AILD), which was ultimately withdrawn and did not become binding, and the Revised Product Liability Directive (R-PLD), which is currently in force. These two instruments will be examined along three key dimensions: (1) the models of liability on which each instrument is based; (2) how each instrument allocates the burdens of proof; and (3) the selection of facts that may be presumed, considering the practical difficulties in establishing causal links in cases involving opaque AI technologies. The lectures in this chapter will conclude with a critique of the limited scope of these instruments, both in terms of the types of disputes to which they are likely to apply and the types of liability excluded from their scope, namely liability arising in the field of defence.
Learning outcomes of the learning unit
Solid understanding of the approaches and methods followed in regulating AI and big data in the EU.
Solid knowledge and understanding of the regulatory challenges as well as opportunities presented by new technologies, in particular AI.
Development of the ability to discuss in English before an audience comprised of lawyers, political scientists and engineers.
Active participation in debates on the six main themes addressed in the seminar.
This course contributes to the learning outcomes II.1, II.2, V.1, V.2, VI.1, VI.2, VI.3, VI.4, VII.1, VII.2, VII.3, VII.4, VII.5 of the MSc in data science and engineering.
This course contributes to the learning outcomes II.1, II.2, V.1, V.2, VI.1, VI.2, VI.3, VI.4, VII.1, VII.2, VII.3, VII.4, VII.5 of the MSc in computer science and engineering.
Prerequisite knowledge and skills
Openeness to, and exploration of various aspects of the interrelationship between law and new technologies
Planned learning activities and teaching methods
The seminar will take place in person, with 24 hours of lectures delivered ex cathedra. For each Chapter, the students will receive a list of sources (textbooks, articles and judgments) with a selection of mandatory readings. The lectures will be interactive, including discussions between the Professor and the students namely on the sources selected as mandatory. PPTs will be used on specific points of the seminar, and will also be made available to the students.
Mode of delivery (face to face, distance learning, hybrid learning)
Face-to-face course
Course materials and recommended or required readings
Relevant materials (presentations, monographs, articles, caselaw) will be made available to the students for each of the four Chapters included in the seminar. A selection of sources for mandatory reading will also be communicated to the students.
Exam(s) in session
Any session
- In-person
written exam ( multiple-choice questionnaire, open-ended questions )
Further information:
By default, assessment takes the form af a closed-book written exam (multiple-choice and open-ended questions). Students are nevertheless permitted to bring the relevant legislation with them.
By way of exception, Erasmus students will take an oral examination (with a preparation period, and on a closed-book basis) instead of the written examination.
By way of exception, students enrolled in one of the programmes listed below will take the oral examination (with a preparation period, and on a closed-book basis) instead of the written examination:
- Master in multilingual communication, professional focus on digital media education
- Master MSc. in Data Science, professional focus
- Master MSc. in Computer Science, professional focus in intelligent systems
- Master MSc. in Computer Science, professional focus in intelligent systems (Double degrees - HEC Liège)
- Master Msc. in computer science and engineering, professional focus in intelligent systems
- Master Msc. in computer science and engineering, professional focus in intelligent systems (Double degrees - HEC Liège)
- Master MSc. in Data Science and Engineering, professional focus
Work placement(s)
Organisational remarks and main changes to the course
See the mode of delivery tab above
Contacts
Lecturer:
Ljupcho Grozdanovski (lgrozdanovski@uliege.be)
Jérôme De Cooman (Jerome.decooman@uliege.be)
Association of one or more MOOCs
There is no MOOC associated with this course.