Admission requirements
None.
Note: The course is of interest also to students from the faculties of Governance and Global Affairs (Institute of Security and Global Affairs), Humanities (Institute for Philosophy), and the School of Law.
Description
Topics
AI & Society: AI, embodied AI, & generative AI; Regulation of AI: Law, standards, science for AI/robot policy; Algorithmic decisions & fairness; Discrimination: diversity, equity & inclusion, inferential analytics, gender bias; Algorithmic speech & freedom of expression; Civil responsibility and liability for AI; Criminal liability; Transparency, explainability, & opacity.
Disciplines
Law, Public Policy, Engineering, Philosophy.
Content description
The integration of artificial intelligence (AI) technologies in society is accelerating. AI-based systems can be software-based, such as image analysis software used for diagnoses, recommendation systems for mortgages, crime, or patient management, search engines, or speech and facial recognition systems used for a wide range of applications, from border control to beauty technologies. AI can also act in the real world embodied in physical support that can range from minimal, such as speakers that support voice assistants like Alexa or Google Home, or more complex hardware devices that perform intricate tasks, such as advanced surgical robots, exoskeletons, self-driving delivery vehicles, flying taxis, or farming robots. Some latest advances include generative AI algorithms (such as ChatGPT or DALL-E) used to create new content, including audio, code, images, text, simulations, and videos.
The High-Level Expert Group on Artificial Intelligence appointed by the European Commission defined AI as 'systems that display intelligent behavior by analyzing their environment and taking actions – with some degree of autonomy – to achieve specific goals.' These technologies process vast amounts of data, can learn from experience, and self-improve their performance, which challenges the applicability of existing regulations that were not designed for progressive and adaptive AI. Since the automated processing of data will evaluate, analyze, and predict outcomes that may affect the privacy, safety, or dignity of individuals, there is a growing interest in understanding the legal and regulatory implications of AI in society.
In this course, we first map the latest advances in AI, including software-based applications and techniques, such as gender classifier systems, generative AI, predictive police, facial recognition, personalized diagnosis, early disease detection, drug discovery, and algorithmic decision-making used in various contexts, including law enforcement and social media. We then identify how the EU approaches these developments from a regulatory viewpoint, mapping the main regulatory initiatives revolving around AI technologies in the European Union, including public policymaking (i.e., AI Act, Machinery Directive, AI Liability Directive) and private standards (i.e., ISO Standards). Since technology and regulation usually evolve at different speeds and directions, often creating dissonances and disconnects, we will also reason about new ways of aligning regulation and practices via science for robot/AI policy.
Since inserting AI and robots in society is not straightforward and challenges many aspects of human life, from how humans interact with each other to how we conceive the world around us, during this course, we will focus on the benefits but also particular challenges revolving around the deployment of AI in society. For instance, we will discuss how false beliefs revolving around apparently neutral and objective computational methods and data may lead to exclusion, inequality, and injustice for vulnerable groups. We also cover the requirements these systems should comply with to be safe and, if damage occurs, the grounds they are liable for and up to what amount. Since these systems could be used directly or indirectly to commit a crime, we will also discuss how criminal liability is being shaped due to advancements in AI. We close the course reflecting on proposed solutions to some of the arisen problems, such as transparency and explainability, and how the law should balance innovation and user rights protection in the context of AI.
Course objectives
The course Law and Artificial Intelligence has five main objectives:
Learn about emerging AI technologies, applications, and benefits.
Identify the main challenges of AI are from legal and regulatory perspectives.
Map the regulatory initiatives revolving around AI in Europe and the world.
Assess the legal and ethical risks posed by AI through different law fields.
Think critically about the deployment of AI in society.
Achievement levels
Upon successful completion of this course, students will have the ability to weigh and evaluate the development of specific AI applications, to see where potential regulatory and ethical challenges might arise in their use or deployment, and to learn methodologies to be able to give advice on designing AI technologies in a way that mitigates such issues. In particular:
Understand the basic architecture and operation of AI and the directions in which it will develop in the near future.
Identify and recognize the potential benefits and drawbacks of the deployment of AI.
Learn and understand the fundamental regulatory issues that have emerged in relation to the deployment of AI, and the relevance of design choices in the architecture of AI.
Understand the complexity of the regulatory and policy landscape to address the legal and regulatory issues arising from the use of AI.
Timetable
Check MyTimetable.
Mode of instruction
Lectures
Number of (2 hour) lectures/seminars: 9. Closing lecture/seminar 4 hours.
Names of lecturers: E. Fosch Villaronga and invited lectures.
Required preparation by students: Reading the required materials on Brightspace, submitting one assignment.
Interactive lectures for which registration is required (in MyStudymap/Usis all interactive lectures are mentioned as working groups).
Attendance on all the dates is not mandatory but each session contains essential information to help you prepare for the final exam and the group work
Assessment method
Examination form(s)
Group work (35%);
Final exam (65%).
The group work assignment is obligatory. The group work assignment counts for 35% of the final grade, must be successfully completed with at least a grade of 5.5., and cannot be retaken.
The written final exam counts for 65% of the final grade. The exam must be completed with at least a grade of 5.5. A lower grade than 5.5 in the written exam will make students go for a retake. Only the final exam can be retaken, whereby the assignments and group work grades remain valid.
If the course is not completed with a passing grade within the academic year, any grades obtained for the group work and its assignments will expire by the end of the academic year.
Submission procedures
To be announced through Brightspace.
Areas to be tested within the exam
The examination syllabus consists of the required reading (literature) for the course, the course information guide and the subjects taught in the lectures, the seminars and all other instructions which are part of the course.
Regulation retake passed exams
In this course it is possible to retake an exam that has been passed (cf. art. 4.1.8 and further of the Course and Examination Regulations) on the condition that this course is not part of the minor. Students who have passed the exam may retake the final written assessment (test) of the course if they meet certain requirements. For more information, go to the website > ‘Law’ tab > ‘Retake a passed exam’.
Reading list
Obligatory course materials
Literature:
- All required readings are available via Brightspace in week-by-week folders
Course information guide:
- See Brightspace
Recommended course materials
- See Brightspace
Registration
Registration for courses and exams takes place via MyStudymap. If you do not have access to MyStudymap (guest students), look here (under the Law-tab) for more information on the registration procedure in your situation.
Exchange students have priority and will be registered for the course first. Any remaining seats will be available for students from Leiden University and other Dutch Universities.
Contact
Coordinator: Eduard Fosch Villaronga
Institution/division
Institute: eLaw Center for Law and Digital Technologies
Department: Interdisciplinary Study of the Law (Metajuridica)
Telephone number secretary: 071 – 527 8838
E-mail: elaw@law.leidenuniv.nl
Remarks
Minor students have priority to follow the course, but they are responsible for the registration in MyStudymap/uSis. If minor students do not register on time, this priority will lapse.