Understand the model
Choose between rules, machine learning, and generative AI. Understand training, inference, data quality, and evaluation.
M2 · Fall 2026 · Faculty of Science
Applied AI for
web & cybersecurity.
Learn how AI works, build useful applications, and evaluate the decisions they make. One shared foundation, with practical work for both disciplines.
01 / WHAT WE WILL DO
Build it. Test it. Explain your choices.
We will move from the fundamentals of machine learning to document assistants, evidence-based research, and controlled AI workflows.
Build the application
Connect models to documents and APIs. Create retrieval-augmented generation (RAG) systems and orchestrate workflows.
Make it trustworthy
Verify sources, test failure cases, protect sensitive data, and decide where human approval belongs.
02 / HOW WE WILL LEARN
A shared lesson. Two ways to apply it.
Each topic starts with a short explanation and a worked example. You will then apply the same technique to a problem in your specialty, compare results, and discuss what failed. Mixed teams will bring both perspectives together.
Build useful AI experiences
Work on semantic search, document assistants, structured outputs, and API integrations. Test usability, reliability, and access controls.
Investigate and secure AI systems
Work on phishing classification, open-source intelligence (OSINT), incident summaries, and alert triage. Test evidence quality, permissions, and abuse cases.
Both tracks will write code, evaluate outputs, and explain design decisions. Ethics and security are part of every assignment. A useful result must be supported by evidence and meaningful tests.
03 / START HERE
Take the benchmark before the lesson.
The 20-question assessment takes about 20 minutes. Select, match, and order your answers—no long written responses. It is ungraded and helps us identify what to cover together.
- Enter your name and select Web or Cyber.
- Complete the questions independently, without AI or notes. “Not sure” is a useful answer.
- Download your JSON result or print it. Follow the instructor’s Teams instructions for submission.
Results are not submitted automatically. Keep your result so you can compare your progress later.
Keep the discussion going.
Use our Teams channel for questions, class announcements, and assignment instructions.