Teacher login →NextBridge AI Lab · Grade 8

Waste Sorter AI Project

Runs in this browser. No login. Nothing is uploaded.

Teacher steps

  1. Problem scoping: waste is mixed in school bins.
  2. Data: add sample pictures or teacher photos of paper, plastic and wet waste (no people).
  3. Model: the machine learns from the examples.
  4. Evaluate: test 10 items, count right and wrong.
  5. Fairness: test unusual items (crushed bottle). What fails?
  6. Responsible AI: write 3 rules for using this tool safely.

Time: 40–80 min · Maps to: CBSE Computational Thinking & AI, Class 8 (AI project cycle, no-code tools, fairness, responsible AI)

Activity

Pictures stay on this device. Nothing is uploaded or saved. Teacher-led: the camera is off until the teacher turns it on. Use objects, not faces.

Low-bandwidth mode: “Use sample pictures” needs no camera and no photos. The AI model is about 2 MB and loads once. No camera? Use sample pictures or add pictures.

📄 Paper (0)

🧴 Plastic (0)

🍌 Wet waste (0)

Test the machine

Levels

  1. Try it: use sample pictures, then test.
  2. Train it: add your own photos of real school waste.
  3. Break it & fix it: leave out one type (e.g. no cups) and test a cup. Fix by adding that data.

Badges:

Badges are saved only in this browser (localStorage). Nothing is sent.

How AI works

  1. The AI project cycle: problem → data → model → evaluate → deploy.
  2. The model can only know what its data showed it.
  3. Testing finds mistakes and unfairness before real use.

What we learned: The AI project cycle: problem → data → model → evaluate → deploy. Testing finds mistakes and unfairness before real use.

Aligned with AI4K12 Big Idea 3: Learning & Big Idea 5: Societal Impact; mapped to CBSE Computational Thinking & AI, Class 8 (AI project cycle, no-code tools, fairness, responsible AI).

Teacher guide

Goal: The AI project cycle: problem → data → model → evaluate → deploy. Time: 40–80 min. CBSE unit: CBSE Computational Thinking & AI, Class 8 (AI project cycle, no-code tools, fairness, responsible AI).

Discussion questions

  1. Which step of the project cycle was hardest?
  2. What would happen if a real bin used a bad model?
  3. Who is responsible if the AI is wrong?

Exit quiz

  1. Name the 5 project-cycle steps.
  2. Accuracy = right answers ÷ ___.
  3. Should student photos be used? (yes/no)

Answers

  1. Name the 5 project-cycle steps. Problem scoping, data acquisition, data exploration/model, evaluation, deployment
  2. Accuracy = right answers ÷ ___. total tests
  3. Should student photos be used? (yes/no) no

Safety: Teacher-led. Use only the teacher's device. Camera is off until the teacher turns it on. Do not show children's faces; use objects. No student personal data. No class codes, dashboards or galleries are used.

Class certificate

Type a first name. The certificate is made on this device and is not saved or sent.

← AI Lab AI practicals

Aligned with NEP 2020 and mapped to CBSE’s Computational Thinking & AI curriculum (Classes 3–8). Big Ideas follow the AI4K12 framework. NextBridge is independent and not affiliated with CBSE or AI4K12.

NextBridge AI Lab · Grade 8 · Waste Sorter AI Project — Worksheet

Name: ______________ Class: ____ Date: ______

  1. Fill in the AI project cycle for waste sorting.
  2. Record 10 tests: item, guess, right/wrong.
  3. Write 3 responsible-AI rules.

Exit quiz:

  1. Name the 5 project-cycle steps.
  2. Accuracy = right answers ÷ ___.
  3. Should student photos be used? (yes/no)

Teacher answer key · Waste Sorter AI Project

  1. Problem → data → model → evaluate → deploy, filled for waste.
  2. Accuracy = right ÷ 10.
  3. e.g. no personal photos; test before use; a human checks.

Exit quiz:

  1. Name the 5 project-cycle steps. Problem scoping, data acquisition, data exploration/model, evaluation, deployment
  2. Accuracy = right answers ÷ ___. total tests
  3. Should student photos be used? (yes/no) no

Maps to: CBSE Computational Thinking & AI, Class 8 (AI project cycle, no-code tools, fairness, responsible AI)

Book a Growth CheckWhatsApp