Regulation and Governance of AI
Course description
These days Artificial Intelligence (AI) is everywhere. AI technologies are quickly becoming
part of a wide range of products, services, and applications, including self-driving cars,
Internet of Things devices in the home, social media, assembly lines, decision-support for
legal systems, risk assessment systems in insurance and law enforcement, and diagnostic
and predictive systems in hospitals and factories. They open up a range of new possibilities
and activities and make it possible for people to connect and interact with each other and
the world in new ways. As such, they challenge our existing norms and ways of regulating
and governing these activities. But what exactly is AI? And what are the particular regulatory
issues that it raises?
What will you learn?
The course will help you to develop a general understanding of what AI is and what the
leading current issues are with regard to regulation and governance. It will also enable you
to identify, explain and look for ways to address legal and ethical issues that developments
in AI may raise.
After a general multidisciplinary introduction, the course explores how developments in AI
challenge existing legal frameworks and what kind of regulatory options can be used to
address these challenges. The national and international dimensions of these challenges will
be discussed, as well as how AI affects different parts of society.
The course consists of three modules:
1. A general introduction to AI and its social and ethical issues
2. A focus on regulation and governance by AI
3. Exploring AI in particular practices through inter-active group-based learning
Learning objectives
1. You will have a general understanding of AI as a field of research and as a technology,
and what sets it apart from other technologies.
2. You are able to identify and explain key regulatory, legal and ethical issues particular
to AI.
1
, 3. You have a general understanding of how developments in AI affect different fields of
law and regulatory paradigms and how legal frameworks challenge developments in
AI.
4. You have general understanding of what governance of and through AI technologies
entails on a local and global level.
5. You are able to critically analyse and discuss current debates about the regulation
and governance of AI.
6. You are able to identify the key issues and concepts that underlie these debates and
communicate about this to an audience of legal scholars.
For the mid-term: objectives 1,2,3,4.
Module 1: General introduction to AI and its social and
ethical issues
Exploring AI
The first module explores AI from a number of different scientific disciplines, including Law,
STS, Ethics, Philosophy and Computer Science, in order to understand the basics of AI as a
necessary foundation to further explore the regulatory, legal and ethical issues of AI. This
introduction will provide you with a general understanding of AI as a field of research and as
a technology. We will discuss the history of AI and give a broad overview of the technologies
that are currently described as AI, including neural networks, deep learning and rule-based
systems. The students will also learn about the problems that research in AI aims to solve
such as Computer Visions, Natural Language Processing, modelling behaviour and
automated reasoning.
In addition, this module will address some of the social and ethical issues that the current
development and use of AI pose.
The module will consist of 4 interactive online lectures (2 weeks):
- Introduction to AI: defining AI, history of AI, and future visions (Merel Noorman)
- AI technologies explained for lawyers – rule based expert systems (Floris Bex)
- AI technologies explained for lawyers – machine learning (Floris Bex)
- Key social and ethical issues of recent AI developments (Merel Noorman)
Module 2: AI from the perspective of regulation and
governance
2
,Another perspective
The second cluster will explore AI from the perspective of regulation and governance. It will
look at how the use of AI technologies for applications - such as face recognition, targeted
advertising, risks assessment, price discrimination and robotics - challenge existing legal
frameworks and regulatory mechanisms, at a local as well as global level. In addition, this
module discusses how these challenges can be addressed and how AI technologies are and
can be used for regulatory purposes.
4 interactive online lectures:
- Regulating AI: From very abstract to very specific (Ronald Leenes)
- Paradigms and problems in AI regulation (Linnet Taylor)
- How AI challenges fundamental regulatory concepts, such as agency, ownership,
responsibility, and personhood (Merel Noorman)
- Technoregulation: AI as a regulatory tool (Ronald Leenes)
Mid-term
Test: It is an open-book take-home test
Dates: 28 April 18:30 – 30 April 18:30
Goals
- The goal of this formative midterm is to assess what you’ve learned in the first 7
lectures. More specifically the midterm will test your understanding of AI and its
multifaceted nature as well as your ability to identify and explain regulatory, legal
and ethical issues particular to AI.
Overview
- 5 essay questions and some smaller questions covering the first seven lectures of the
course
- 4 essay question in 15-20 minutes.
- 1 essay question +/- 45 minutes.
- Smaller questions will take a few minutes to answer
- Allowed to use all the reading material, video clips, and slides.
AI regulation proposal part of the midterm? There is no direct question on the regulation,
but it can be part of the answer. You should get acquainted with the draft regulation.
In the description it is mentioned “more specifically the midterm will test […] your ability to
identify and explain regulatory, legal and ethical issues particular to AI”.
3
, Could you please clarify whether strictly technical questions will be included (such as
building algorithms as we did in the assignments)?
There will be no technical questions in the sense of build an algorithm, but as it says in
the description you will be tested on your understanding of AI, and that includes
understanding some of the basic technical aspects as discussed in the two lectures. So, no
practical exercise, but there will be question on the technical aspects.
4