AI Recycling Sorter Audit
Log in to saveKids run a simple image-based sorting tool on pictures of recyclables and compare its calls against a human-made sorting rulebook, digging into where and why the two disagree. It's an approachable, hands-on introduction to how classification systems work and where they fail.
Choose your version
Every level teaches the same core idea — pick the one that fits your time, budget, and energy today.
Setup
Place common or recycled supplies for AI Recycling Sorter Audit on one work surface and choose a safe area for the test. Read the four activity directions aloud; drawings, paper models, and household substitutes may replace specialized materials, and no digital tool is required.
Activity
- Look at a set of item pictures and predict how a simple sorting rulebook would classify each one as recyclable or not.
- Build the sorting stations and set up both the human rulebook and the simple automated sorter for comparison.
- Test both systems on the same set of items, then investigate and adjust the rulebook when the two sorters disagree on a tricky item.
- Present the audit results to the group, explaining which items were hardest to classify and why the systems disagreed.
Done when: The group produces the AI Recycling Sorter Audit solution, completes the described test, makes the revision called for by the evidence, and supports its final explanation with an observed or measured result.
Materials
printed pictures of recyclable and non-recyclable items, recycling sorting bins, printed AI rulebook cards, clipboard for tallying
Shown for the Quick Start version — switch tabs above to see what changes.
Steps
- Look at a set of item pictures and predict how a simple sorting rulebook would classify each one as recyclable or not.
- Build the sorting stations and set up both the human rulebook and the simple automated sorter for comparison.
- Test both systems on the same set of items, then investigate and adjust the rulebook when the two sorters disagree on a tricky item.
- Present the audit results to the group, explaining which items were hardest to classify and why the systems disagreed.