Stage 1 · Lesson 6 of 8
Why models hallucinate
A is a confident answer that is false. Think of a made-up citation, an method that does not exist, or a company policy that was never written. It comes from how models work.
They predict likely text, and a plausible-sounding answer is often more likely than "I do not know". This is especially true for rare facts, or anything after the (the date the model's training data ends). You cannot prompt it away entirely, but you can shrink it.
Give the model the source text. Ask it to answer only from that text and to cite it. Give it an explicit way to refuse.
Check claims in code where you can. This is the problem and exist to manage.
A question to check yourself, an AI practice session and 1 cheat sheet entry come with the full roadmap. Unlock
Resources for this stage
Start with· follow one of these
- CoursePaidGrokking Modern AI Fundamentals
Design Gurus · Beginner · $72 one-time
Structured grounding in models, transformers and embeddings before you build on them
- VideoFreeDeep Dive into Large Language Models like ChatGPT
Andrej Karpathy · Intermediate
Free: Karpathy on tokens, training and why models hallucinate, in one sitting
Keep handy
Also worth your time
Input vs output tokens and a step by step cost estimate you can reuse for any feature
- Neural Networks: Zero to HeroGo deeper
Build a small GPT from scratch if you want to see the machinery for yourself