AI-Augmented Curriculum Adaptation and Prompt Design Guide

Act as an experienced curriculum designer and responsible AI-in-education specialist.

Develop an AI-augmented curriculum adaptation and prompt design guide using the information below.

Curriculum or course:
{{curriculum_context}}

Adaptation need:
{{adaptation_need}}

Learning outcomes and non-negotiables:
{{learning_outcomes}}

Source materials:
{{source_materials}}

Learner and accessibility context:
{{learner_context}}

AI tools, policies, and constraints:
{{ai_policy_context}}

Design requirements:

1. Define the adaptation goal precisely.
2. Identify what must remain unchanged, including:
   - learning outcomes;
   - disciplinary accuracy;
   - assessment construct;
   - required standards;
   - legal or regulatory content;
   - essential terminology; and
   - institutional policy.
3. Identify appropriate AI-supported tasks, such as:
   - reading-level adjustment;
   - example generation;
   - question generation;
   - scenario variation;
   - language support;
   - formatting;
   - summary;
   - alternative explanation;
   - feedback drafting;
   - metadata;
   - accessibility description; and
   - localization.
4. Identify tasks that require expert human judgment or should not be delegated.
5. Create a prompt template containing:
   - role;
   - context;
   - source material;
   - learning objective;
   - target learners;
   - adaptation task;
   - constraints;
   - required output;
   - quality criteria;
   - uncertainty instruction; and
   - verification step.
6. Create specialized prompt patterns for at least four relevant adaptation tasks.
7. Require the AI to preserve source meaning, flag uncertainty, and avoid invented facts or citations.
8. Include a human-review workflow covering:
   - subject accuracy;
   - pedagogy;
   - bias;
   - cultural relevance;
   - accessibility;
   - copyright;
   - privacy;
   - academic integrity;
   - assessment validity; and
   - policy compliance.
9. Define handling rules for confidential, student, copyrighted, proprietary, and sensitive information.
10. Create a version-control and change-log procedure.
11. Include pilot testing with educators and learners where appropriate.
12. Define quality indicators and failure conditions.
13. Do not assume that AI output is accurate, original, unbiased, accessible, or policy compliant.
14. Do not upload protected data unless authorized and appropriately safeguarded.
15. Flag decisions requiring legal, privacy, accessibility, assessment, IT, or institutional approval.

Present the result as:
{{output_format}}

Include:
- adaptation objective;
- non-negotiable elements;
- appropriate and inappropriate AI tasks;
- master prompt template;
- task-specific prompt patterns;
- human-review workflow;
- quality checklist;
- privacy and copyright safeguards;
- accessibility and bias review;
- version-control process;
- pilot plan;
- failure and escalation rules; and
- implementation guidance.