AI Guessing Game
A possible investigation of pattern-based guesses, confidence, errors, and fair testing.
See the later ideaPossible later program / AI & Smart Machines
A future program could investigate how AI tools produce predictions, text, and images—and how confidently those tools can be wrong.
Ask Eugene About AIPractical questions
Students can compare prompts, inspect patterns, test outputs, collect confident mistakes, and distinguish a useful prediction from understanding.
The central habit is verification: what evidence supports the output, what may be missing, who could be affected, and what a person remains responsible for deciding.
Possible topic path
Two future previews
A possible investigation of pattern-based guesses, confidence, errors, and fair testing.
See the later ideaA possible writing and testing lab about prompts, character rules, inconsistent answers, and verification.
See the later ideaClass and Computer Lab
A class would set a bounded question, establish privacy rules, inspect examples, and compare results. Computer Lab work could design harder tests, classify failures, revise a prompt, or connect a limited prediction to a game or robot rule.
Fluent language is not evidence. A confident answer can still be wrong.
Human judgment remains responsible.
Any future AI work will state which tools are used, what data they receive, what claims are being tested, and which decisions remain with a person.
Read the AI parent guideAsk Eugene about the idea. Scratch and Robotics are the programs available now.