AdmiScan Research Note
Cloud Came. Docker Came. AI Came. What Should Students Learn Next?
Fifteen years ago, cloud computing sounded like a specialist technology. Then came containers, Docker and Kubernetes. Today, artificial intelligence dominates technology conversations. But students entering university in 2027 may not graduate until 2030 or later. So what should they be learning now?
Research status: ongoing
AdmiScan is building a longitudinal dataset from official university sources. This page documents the research question, method and source base. It does not claim that a single technology has already been identified as “the next AI”.
The question
Technology education has always moved in overlapping waves. Universities adapted to web development, mobile computing, cloud platforms and distributed systems. Later, virtualisation and containers became increasingly important. Today, machine learning, generative AI and large language models are rapidly entering computer-science classrooms.
The more useful question is not whether AI replaces those technologies. It is which technologies universities are beginning to teach now that could become mainstream by the time today’s students graduate.
How the study works
AdmiScan is comparing official curricula across three broad periods: 2012–2015, 2019–2022 and 2025–2027. Historical absence is not treated as proof that a subject was not taught. The analysis records verifiable positive evidence and uses archived catalogues only when they are sufficiently complete for comparison.
Official-source pilot
The initial feasibility work has locked historical and current official-source pairs for eight universities in the United States, United Kingdom and Germany. These are being used to test the longitudinal method before the dataset is expanded.
| University | Historical evidence | Current evidence |
|---|---|---|
| Stanford University | 2014–15 Computer Science schedule | Current Computer Science MS specialisations |
| Carnegie Mellon University | 2014 Cloud Computing syllabus | Current Generative AI / ML systems teaching |
| MIT | 2014 EECS teaching record | AI and Decision Making track |
| UC Berkeley | Archived CS188 Artificial Intelligence material | Current CS188 catalogue |
| University of Oxford | 2013–14 Machine Learning course | Current Computer Science course catalogue |
| University of Cambridge | 2013–14 Computer Science Tripos archive | Principles of Machine Learning Systems |
| University of Edinburgh | 2013–14 Informatics archive | Current Informatics curriculum |
| University of Bamberg | Archived Software Systems Science module handbook | AI & Data Science module handbook |
An early signal worth testing
One early pattern worth testing is the growth of engineering around AI itself: machine-learning systems, distributed training, scalable inference and AI infrastructure. Robotics, autonomous systems, cybersecurity, quantum computing and responsible AI also appear in current curricula.
The expanded dataset will determine whether any of these are broad curriculum shifts or simply institution-specific examples. AdmiScan will not label a technology “the next big thing” unless the evidence supports it.
Related analysis
If AI changes the nature of work itself, students face a second question: what should education prepare them for?
Read: If work becomes optional, what should students study for?