Learning Targets:

1. Configure a model with an appropriate input type (image, sound, or text) and select sensible parameters provided by the tool.

2. Execute training and record key settings (e.g., number of samples per class, epochs if shown).

3. Evaluate model performance on the test set using accuracy and, where available, a confusion matrix or precision/recall.

4. Conduct error analysis by inspecting misclassified examples and identifying likely causes (e.g., noisy labels, class imbalance).

5. Iterate by improving data quality or adjusting parameters to achieve a measurable performance gain (e.g., at least a 5% accuracy increase or fewer false positives).

Modules

Learning Targets:

1. Structure prompts with a clear role, task, constraints, audience, and examples (R-T-C-A-E) to guide outputs.

2. Apply iterative prompting by refining at least two drafts based on specific feedback criteria (clarity, tone, length, evidence).

3. Fact-check at least three claims produced by the AI using trusted sources and provide simple citations.

4. Red-team prompts to avoid unsafe, biased, or inappropriate outputs and revise wording to mitigate risks.

5. Compare outputs under different settings (e.g., temperature or style options where available) and note effects on quality and reliability.

Modules

Learning Targets:

1. Identify potential harms (privacy breaches, stereotyping, misinformation, academic integrity) within a given AI scenario.

2. Evaluate a dataset or prompt for fairness by checking class balance, stereotype wording, and representation.

3. Propose at least three mitigations (e.g., improve data balance, add safety instructions to prompts, secure storage) and justify each choice.

4. Articulate clear usage and citation guidelines for classroom AI activities that respect intellectual property and school rules.

5. Demonstrate safe platform use by enabling safety features, reporting problematic content, and protecting personal data.

Modules

Learning Targets:

1. Define a problem statement, target users, and success criteria linked to a realistic context (school or community).

2. Select an appropriate AI approach (classifier with Teachable Machine/ML for Kids or a generative AI prompt workflow) and justify the choice.

3. Build a working prototype and maintain a project log capturing decisions, versions, and obstacles.

4. Test the solution with at least five test cases or two user feedback sessions and record results against success criteria.

5. Present the solution using visuals or a short demo, and reflect on impact, limitations, ethics, and next steps.

Modules