Applied AI

Course map · S1 How LLMs work · Episode 9 of 12

Hallucination

S1 E9 Explainer · 22 slides · 6 October 2026

  1. Slide 1 of 22, Cover.
  2. Slide 2 of 22, Try this.
  3. Slide 3 of 22, Why it matters.
  4. Slide 4 of 22, Mental model.
  5. Slide 5 of 22, Concepts.
  6. Slide 6 of 22, Concepts.
  7. Slide 7 of 22, The experiment.
  8. Slide 8 of 22, The setup.
  9. Slide 9 of 22, Data.
  10. Slide 10 of 22, Where it breaks.
  11. Slide 11 of 22, Where it breaks.
  12. Slide 12 of 22, Data.
  13. Slide 13 of 22, Read this with the result.
  14. Slide 14 of 22, For engineers.
  15. Slide 15 of 22, Recap.
  16. Slide 16 of 22, Where it breaks.
  17. Slide 17 of 22, Concepts.
  18. Slide 18 of 22, For engineers.
  19. Slide 19 of 22, In practice.
  20. Slide 20 of 22, For leaders.
  21. Slide 21 of 22, Decision rule.
  22. Slide 22 of 22, Series map.

Lesson notes

Fluent answers can be confidently wrong, and the only fabrication in this test came from the keyword scoring, not the model, so check error rates by reading answers.

Decision rule

Never ship an answer you cannot trace: ground it in a source, let the model say it does not know, and measure the error rate by reading, not just by keyword.

Ideas in this lesson

  • Probability
  • Grounding
  • Evaluation
  • Measuring, not quoting

Take it further