Critical Perspectives on AI2023HT
No of lectures
6-8 seminars (exact number depends on number of participants)
The course is mainly intended for PhD students in cognitive science, computer science, and related disciplines (but see also Prerequisites below).
The course was last given
The course was last given spring 2022.
The main goal is to familiarize students with critical perspectives addressing limitations, risks, misperceptions, etc. of AI research and technology.
Some background in AI, cognitive science, and/or human-computer interaction. Students do not necessarily need much technology/computing background, though, so PhD students who have a strong interest in AI, but come from other research areas (e.g., science & technology studies, gender studies, or applications of AI in education, healthcare, etc.), are also welcome.
The course mainly consists of discussion seminars and student presentations.
The course consists of:
- one introductory lecture/seminar that goes through 'old' criticisms of AI discussed in the 1960s-90s (e.g., frame problem, common sense problem, symbol grounding problem), and
- 5-7 seminars discussing 5-7 recent books (2019-2023) that address critical perspectives on AI (see Literature for details).
The course literature mainly consists of 5-7 of the following books:
- Aylett & Vargas (2021). Living with Robots: What every anxious human needs to know. MIT Press.
- Christian (2020). The Alignment Problem - Machine Learning and Human Values. Norton & Company
- Crawford (2021). Atlas of AI: Power, politics, and the planetary costs of artificial intelligence. Yale University Press.
- Larson (2021). The Myth of Artificial Intelligence: Why computers can't think the way we do. Harvard University Press.
- Russell (2019). Human Compatible: Artificial intelligence and the problem of control. Viking Press.
- Smith (2019). The Promise of Artificial Intelligence: Reckoning and judgement. MIT Press.
- Strengers & Kennedy (2020). The Smart Wife: Why Siri, Alexa, and other smart home devices need a feminist reboot. MIT Press.
- Zweig (2022). Awkward Intelligence: Where AI goes wrong, why it matters, and what we can do about it. MIT Press.
NB: The list above is updated from the spring 2022 edition of the course and might be updated/extended a bit further if relevant new books appear in late 2022 or early 2023.
Mandatory student presentations, active participation in seminar discussions, and some written coursework.
The course can be given in Zoom if there are non-local participants.
Page responsible: Director of Graduate Studies