Welcome to Chapter 7! Before building AI solutions, you need to understand how modern AI systems like ChatGPT actually work — and why they sometimes get things badly wrong.
This chapter runs over 3 weeks: Week 1 covers how LLMs work and why they fail;
Week 2 is fully hands-on — you'll test real AI tools and explore AI careers;
and Week 3 is your exam.
What You'll Learn
Week 1: Tokens, training data, context windows, hallucinations, and the key limitations of LLMs
Week 2: Test a real AI tool and document what you find; explore AI careers and how AI will affect your field
Week 3: 20-question MCQ exam covering all concepts from Weeks 1 & 2
The 3-Week Plan
🔤
Week 1
How LLMs Work & Why They Fail
Learn how text becomes tokens, how LLMs are trained, what context windows are, and why AI confidently makes things up. The theory behind everything.
20 multiple-choice questions across 6 sections. Tests your understanding of LLM concepts, hallucinations, context windows, limitations, and real-world application.
Large Language Models don't "think" — they predict the next token based on patterns learned from data.
Understanding this, plus knowing when and how AI fails, changes how you use and build AI forever.