History of AI
Historical development of computing and AI: from punched cards to stochastic parrots, exploring key milestones, paradigm shifts, and the societal impact of digital culture.
Seminars and workgroups are not mandatory but strongly recommended. There are three types of sessions you can enroll in: lecture recaps (TA-led review of previous lecture), book recap & assignment working sessions, and assignment feedback sessions. Attending seminars helps you absorb the material and prepare for the essay-style exam.
Learning objectives
Students acquire knowledge and understanding about the history of computing / AI from various perspectives: computing as a scientific goal, computing as a government (administrative or military) objective / ideal, computing as an economic enterprise (Knowledge and understanding). Students can apply that knowledge and understanding to describe the meanings digital culture has or may have for various people in contemporary society (Applying knowledge and understanding). Students are able to communicate their knowledge and understanding in an effective way, and from various perspectives — thereby they will be better equipped to reflect on their subject of study (Communication)(Making judgements).
History of AI (also known as "History of Science") takes you on a fascinating journey through the evolution of computing and artificial intelligence, told from historical, philosophical, ethical, and sociological perspectives. Taught by Professor Danny Beckers, the course is built around the textbook "Computer: A History of the Information Machine" by Martin Campbell-Kelly (4th edition, 2023) — chapters 4 through 15. The lectures don't just repeat the book; they run a parallel narrative with a strong emphasis on the European side of the story, while the book focuses more on the American perspective.
The seven lectures cover: (1) the prehistory of AI and computing — from Charles Babbage and punched card machines to Alonzo Church and Alan Turing's foundational work, (2) creating computers — the construction of early machines like EDVAC and the birth of stored-program computing, (3) the sound of software — the rise of programming and software as a discipline, with connections to early sci-fi like Metropolis and Asimov, (4) computing crisis — the software crisis of the 1960s, the evolution of programming languages including ALGOL and LISP, and figures like Dijkstra, (5) dream machines — the space race, early AI programs like SHRDLU, personal computing pioneers like Douglas Engelbart, and the rise of the PC, (6) neat networks — networking, the personal computing boom, and the internet era, and (7) the digital divide — modern computing, algorithmic bias (Timnit Gebru's "On the Dangers of Stochastic Parrots"), surveillance capitalism, and AI ethics.
The course is explicitly reflective rather than prescriptive — the goal is not to memorize dates, but to understand how computing and AI emerged within their social context, to distinguish reality from hype, and to confront your own assumptions about technology. The workload is approximately 40 hours of reading, 14 hours of lectures, and 9 hours working on assignments, leaving about 20 hours for exam preparation. There are six weekly group assignments written in essay format, and the exam features three categories of questions: A-questions (lecture content), B-questions (book content), and C-questions (combining both). You pick four questions to answer, each worth 2 points.
Assessment
Written exam (2h 15min); the exam has 8 A-questions (about lectures), 12 B-questions (about the book), 6 C-questions (combining both), and a NO BONUS question. You pick 4 questions: 1 A-question, 2 B-questions, and 1 C-question, each worth 2 points. If you are not entitled to the bonus point, you must also answer the NO BONUS question (worth 1 point). The bonus is earned if the average of your four best assignments is 6 or higher — this lets you skip the NO BONUS question, effectively starting your exam with 2 points instead of 1. Resit is possible for the exam; resubmission of assignments is generally not allowed.
Teaching methods
Weekly lectures by Professor Beckers; TA-led lecture recap sessions (attending one is enough); book recap & assignment working sessions; assignment feedback sessions. Workgroups and seminars are not mandatory but highly recommended. Optional Canvas quizzes help with studying.
Literature
Martin Campbell-Kelly a.o., Computer: A history of the information machine (fourth edition, 2023). Available as e-book through the VU university library. Students are required to read chapters 4-15. Lectures also reference primary sources such as Turing's "On Computable Numbers" (1936), von Neumann's first draft of EDVAC, Dijkstra's "A Case Against the GOTO Statement", and Gebru et al.'s "On the Dangers of Stochastic Parrots" (2021).