Module 01 / Beginners
AI
Basics.
What is artificial intelligence really? How does an algorithm learn? And where do you already meet it in everyday life? This module gives you a clear, honest introduction, without jargon and without hype.
What is artificial intelligence?
AI is not consciousness, not a brain, not a robot from the movies. AI is a tool, a very powerful one, that learns from data, recognises patterns and supports decisions. Four core ideas you need to know:
Recognising patterns
AI looks for repetition in huge amounts of data. Faces, texts, sounds: it finds structures humans would miss.
Learning from data
AI does not program itself. It is trained. Millions of examples show it what is "right" until it draws conclusions on its own.
Scaling decisions
What a human analyses in an hour, AI does in milliseconds, a thousand times at once.
No "understanding"
AI does not understand what it is doing. It finds correlations. Whether they make sense is still up to us.
Short definition: Artificial intelligence refers to systems that can solve tasks that would otherwise require human intelligence, by learning from examples rather than through explicit programming instructions.
How does an AI learn?
The core of AI is machine learning: a cycle of data, training, evaluation and improvement. The loop repeats millions of times until the model is good enough.
Image recognition: The model is shown millions of photos: dogs, cats, cars, faces, each with a label. It learns which pixel patterns belong to which object. After training it recognises new images it has never seen. Error → adjust weights → repeat.
Text translation: The model is shown millions of sentence pairs (German/English). It learns which word sequences correspond. Over time it gets so good that it can translate sentences that were not in the training set, because it has learned the structure of the language, not just memorised words.
Recommendations: Netflix, Spotify, TikTok: they all use the same principle. The model watches what you click, skip and linger on. It looks for users who resemble you. What they like, you might like. No human decides this: the algorithm infers it from your behaviour.
AI is all around you.
Tap a card to see whether and how AI is involved.
Tap to reveal ↓
Netflix recommendation
You click "For you"
Google Maps route
Calculating the fastest route
Spam filter
The email lands in junk
Spotify autoplay
The next song plays automatically
Face recognition
Phone unlocks with your face
Autocorrect
Typos get corrected
What AI can do.
What it can't do.
- 🚀Recognise millions of patterns in seconds without getting tired
- 📏Apply the same rules consistently, every time, regardless of mood
- 📡Find connections in huge datasets that humans overlook
- ⚙️Do the same task for billions of people at once
- 🎨Generate texts, images and sounds that seem plausible and useful
- 🧩Truly understand meaning: it finds correlations, not causation
- 🤔Common sense: obvious things outside its training context
- ⚖️Make real ethical judgements: it only reflects the data back
- 🌱Be creative out of nothing: it interpolates between what it knows
- 🔮Predict the future when it differs structurally from the past
Important: AI is not a neutral tool. It reflects the data it was trained on, including its biases, gaps and historical injustices. Whoever deploys AI is responsible for the training data.
AI is only as smart as the dataReformRebell / AI basics
we feed it.
Quick exercises.
Choose the statement that does not apply to machine learning:
Describe two situations from your everyday life where you probably come into contact with AI, even if you are not sure. What happens there? What data is used?
For each statement, tap True or False:
AI can independently discover new laws of physics it has never seen.
An AI model can deliver worse results if it was trained on one-sided data.
AI is conscious: it knows what it is doing and can make moral decisions.
The same AI technology can both help and harm, depending on how it is used.