Responsible AI & Data Ethics

Course goals

Upon completion of the course, students

• can identify moral values in technology
• have knowledge of relevant regulation, guidelines and instruments
• have knowledge of the practical application of testing models
• are able to use impact assessment, audits, and engage in critical reflection

Assessment
Students are assessed as follows:

• written exam on theory (30% of the final grade)
• assignments (30%)
• project assignment: evaluation (40%)

To qualify for a repair of the final result the mark needs to be at least a 4, or “AANV”.

Content

This course teaches students to understand how data projects can carry and affect societal values. They understand the socio-technical quality of data and AI in order to reflect on the societal impact of data science practices.
They learn to apply different methods of impact assessments and audits of algorithms, as well as developing value-sensitive design of data projects.
The students get acquainted with the relevant regulation, guidelines and ethical frameworks for data science and AI.

This course enables students to develop responsible data science practices through practical application in a project, and teaches students to carry out dialogic deliberation of value-sensitive design, assess social impact through a formal assessment, and evaluate results through an audit.

The course caters to the development of very much needed knowledge for areas that are labelled as responsible or ethical AI, explainable AI, fair data practices or responsible data science. It includes an introduction to the ethical, legal and social challenges constituted in data practices (e.g. Kitchin 2022). It introduces relevant legislation, guidelines, code of conducts for ethical data practices, and introduces current methods for algorithm design and testing. 

In the applied part, the course focusses on developing a responsible data project, and reviewing the design of data projects through impact assessments and practical testing of datasets, models and variables.

Course form

• lectures
• tutorials on specific aspects of responsible data practices (e.g. GDPR, AI Act, value-sensitive design, etc.),
• individual and group assignments,
• practical (e.g. testing models, auditing algorithms, etcetera)

The main teaching method is project-based.

Additional information

Enrolment in this course is only available by using the pre-enrolment form that will be forwarded to all ADSM students in the second week of September.