Overview
This week you completed the technical arc of the AI content: reinforcement learning through hill climbing and genetic algorithms, unsupervised learning and dimensionality reduction, and finally large language models — their architecture, their training process, their genuine capabilities, and their systematic failure modes. Neural networks, covered in Week 6, provide the technical foundation underneath all of it.
Week 7 content extends the picture from Week 6 into territory that is directly relevant to your students' daily lives. The systems that drive recommendation feeds use reinforcement learning. The platforms that shape what students see and believe use unsupervised clustering at scale. LLMs are tools many of them already use for schoolwork. The pedagogical challenge at every grade level is similar: moving students from a naive relationship with these systems — they work, so use them — to an informed one — here is what they are doing, here is where they fail, here is how to use them well.
The pages linked below address that challenge by grade band. Find yours and read it before your small group discussion.
Choose Your Grade Band
Elementary — Teaching AI Techniques in K–5
Emphasis on pattern recognition as a concept children already understand, accessible analogies for how networks learn, and age-appropriate conversations about AI in the tools students already use.
Middle School — Teaching AI Techniques in Grades 6–8
Emphasis on making neural networks concrete through physical analogies, addressing the specific misconceptions middle schoolers have about LLMs, and building the critical evaluation habits students need to use AI tools wisely.
High School — Teaching AI Techniques in Grades 9–12
Emphasis on technical depth with the LLM training pipeline at full detail, prompt engineering as a teachable skill, and connections to AP courses and emerging professional contexts.