Plain-language summary
What this guide covers
Teachers plan lessons, guide students, explain ideas, assess learning, communicate with families, and adapt instruction. AI may help brainstorm lessons, draft practice questions, suggest supports, outline rubrics, prepare feedback language, translate simple messages, write administrative drafts, and support professional learning. It should not replace teacher judgment or student agency. Do not upload student records into unapproved tools, and do not delegate grading, discipline, safeguarding, disability accommodations, or other consequential decisions to AI.
Teaching is relational, ethical, and context-rich. A tool that can produce a polished worksheet may still be wrong, biased, inaccessible, age-inappropriate, or misaligned with the curriculum. Teachers know classroom context, student needs, family communication norms, school policy, assessment goals, and what students have actually learned. AI can reduce preparation friction, but the teacher remains responsible for accuracy, pedagogy, privacy, accessibility, and whether an activity supports learning.
What you will learn
- Identify teaching tasks where AI can assist with planning, drafting, translation support, and professional learning.
- Separate safe drafting support from high-risk uses involving student records, grading, discipline, safeguarding, disability accommodations, and assessment validity.
- Use a task map to choose review levels for lesson brainstorming, differentiation, rubrics, practice questions, feedback preparation, and administrative writing.
- Create checkpoints for privacy, accuracy, age appropriateness, accessibility, bias, copyright, disclosure, and academic integrity.
- Run a low-risk first-week experiment with approved tools and no student records.
Guide section
Why the role matters and how AI may change tasks
AI can help teachers prepare materials, but teaching is not just material production. A safe workflow protects learning goals, student privacy, institutional rules, and teacher judgment.
O*NET describes secondary teachers as planning lessons, adapting materials, evaluating student progress, maintaining records, communicating with families and staff, and supporting students with different needs. The U.S. Bureau of Labor Statistics describes high school teachers in a U.S. occupational context and reports 2024 wage and employment information with 2024 to 2034 projections. Those sources describe occupations, not individual outcomes. AI may change tasks by helping teachers create first drafts, compare reading levels, produce examples, outline rubrics, and prepare communication. That is different from reliable adoption in a school or district. A model can generate text, but a teacher must decide whether it fits the curriculum, students, age group, classroom culture, assessment purpose, accessibility needs, and institutional policy.
For education, separate pedagogy evidence from product claims. A product may claim to personalize learning or speed lesson planning, but a teacher still needs to ask whether the activity supports the learning objective, preserves student agency, gives accurate information, and keeps assessment fair. UNESCO’s guidance on generative AI in education and research emphasizes policy, human capacity, data protection, and responsible use. In the United States, FERPA is a federal student privacy law; other privacy, accessibility, copyright, and education rules vary by jurisdiction and institution. This guide is general education, not compliance advice.
Guide section
Teacher task map
Use this task map to decide where AI should draft, suggest, organize, translate, or stay out of the workflow.
Task map
| Task or workflow | Possible AI contribution | Human responsibility | Risk level or review requirement |
|---|---|---|---|
| Lesson brainstorming | Suggest hooks, examples, discussion questions, demonstrations, and activities. | Align to standards, curriculum, age level, classroom context, and available time. | Medium review. Check accuracy, bias, age appropriateness, accessibility, and source quality. |
| Differentiation drafts | Draft simpler explanations, extension questions, vocabulary supports, or practice variations. | Adapt based on actual student needs and approved supports without exposing student records. | High review. Do not delegate disability accommodations or individualized decisions to AI. |
| Rubric drafting | Generate a first-draft rubric structure or criteria list. | Confirm learning objectives, assessment validity, student clarity, fairness, and grading policy. | High review. Teacher or institution owns final rubric and grading decisions. |
| Practice questions | Create draft examples, retrieval questions, or exit-ticket prompts. | Verify answers, difficulty, alignment, reading level, and whether questions measure the intended skill. | Medium review. High review for graded or high-stakes assessments. |
| Feedback preparation | Suggest feedback sentence stems or common misconception explanations. | Use teacher knowledge of the student’s work; do not upload student records to unapproved tools. | High review. AI should not grade or determine consequences. |
| Translation support | Draft a simple translation or plain-language version of a non-sensitive message. | Confirm meaning with qualified support when stakes are high and follow family communication policy. | Medium to high review. Do not use for legal, disciplinary, medical, accommodation, or safeguarding messages without approved process. |
| Administrative writing | Draft newsletters, meeting agendas, supply lists, and non-sensitive announcements. | Review facts, dates, tone, accessibility, approval, and institutional policy before sending. | Medium review. High review if student-specific information appears. |
| Professional learning | Summarize public articles, generate reflection questions, or compare teaching strategies. | Check source quality, research basis, local fit, and whether recommendations match the learning goal. | Medium review. Do not treat AI summaries as research proof. |
Guide section
Good starting tasks and unsuitable uses
Start with teacher-facing preparation that uses no student records and ends with teacher review.
Lower-risk starting tasks
- Brainstorm three lesson hooks for a topic using only the public curriculum objective.
- Rewrite a teacher-created explanation at a lower reading level, then check for lost meaning.
- Draft five ungraded practice questions and verify the answers yourself.
- Create a first-draft rubric outline for teacher editing before sharing with students.
- Generate a classroom discussion protocol or reflection prompt for a non-sensitive topic.
- Summarize a public professional-learning article and compare the summary with the original source.
- Draft a general family newsletter with no student-specific information.
- Create an accessibility checklist for handouts, slides, links, captions, and reading order.
Unsuitable, sensitive, or high-risk uses
- Uploading student records, grades, behavior notes, disability information, family contact details, or identifiable student work into unapproved tools.
- Delegating grading, promotion, discipline, safeguarding, disability accommodations, placement, or other consequential decisions to AI.
- Using AI-generated materials without checking facts, sources, reading level, bias, cultural context, and age appropriateness.
- Creating high-stakes assessments without teacher and institutional review for validity, fairness, security, and academic integrity.
- Using AI to write family messages about discipline, health, disability services, legal rights, or safety without approved human process.
- Presenting AI-generated text, images, or translations as human-created or professionally verified when they are not.
- Copying copyrighted materials into AI tools or publishing AI outputs without checking copyright, licensing, and attribution policy.
- Letting AI replace student thinking by giving answers where the learning goal is reasoning, writing, practice, or explanation.
Guide section
Hypothetical workflow: draft a practice activity
This example is hypothetical. It uses AI to prepare a teacher-reviewed practice activity, not to grade students or decide accommodations.
Example
Inputs and outputs
Inputs: public learning objective, teacher-created notes, grade band, time limit, allowed vocabulary list, accessibility requirements, and school-approved AI policy. Outputs: teacher-reviewed practice questions, answer key, accessibility checklist, teacher notes, and a list of items needing human review. No student records or identifiable student work are used.
Workflow steps
- Confirm the tool is approved for teacher planning and that no student records will be entered.
- Write the learning objective, grade band, time limit, and what skill the practice should measure.
- Ask AI for a draft set of practice questions, an answer key, and a note explaining the reasoning for each answer.
- Check every question for accuracy, curriculum alignment, reading level, age appropriateness, bias, and accessibility.
- Revise the questions so they support student thinking rather than answer-copying.
- Compare the answer key with trusted sources and teacher knowledge.
- Label the final material as teacher-reviewed and note any allowed AI use according to school policy.
- Use classroom results only through normal teacher observation and assessment processes, not through unapproved AI analysis of student records.
Reusable prompt for practice questions
Create a draft practice activity for **{{grade_band}}** students learning **{{learning_objective}}**. Use only the information provided here. Do not include student data. Include **{{number_of_questions}}** practice questions, a draft answer key, likely misconceptions, and accessibility checks. Mark uncertain facts as **Needs teacher review**. Keep the language age-appropriate and avoid stereotypes.Editable fields: grade_band, learning_objective, number_of_questions
Guide section
Human checkpoints, escalation triggers, stop conditions, and ownership
The teacher owns the learning design unless the institution assigns another authorized reviewer. AI should not create hidden decisions about students.
Human checkpoints
- Teacher decision owner: final lesson design, classroom use, feedback wording, rubric approval, and whether material fits the learning goal.
- Institution decision owner: approved tools, student-data rules, disclosure rules, records, copyright policy, and assessment policy.
- Before using AI: confirm the tool is approved and remove student records unless an approved system and policy explicitly allow the use.
- Before sharing materials: check accuracy, bias, reading level, accessibility, age appropriateness, source quality, copyright, and curriculum alignment.
- Before assessment use: verify the task measures the intended skill and does not weaken academic integrity.
- Before translation use: confirm meaning and follow institutional policy for family communication.
Escalation triggers and stop conditions
- Stop if the task requires student records, grades, disability information, health information, behavior reports, safeguarding details, or family contact information in an unapproved tool.
- Escalate if the material affects grading, promotion, discipline, accommodations, safety, legal rights, or official records.
- Stop if the AI output includes stereotypes, unsafe content, false facts, uncited claims, inappropriate reading level, or copyrighted content that cannot be cleared.
- Escalate if a student, family, or colleague asks how AI was used and institutional disclosure rules are unclear.
- Stop if AI use would make it harder to tell what a student actually knows or can do.
- Escalate if translation or accessibility needs require qualified human support.
Guide section
Skills to build, first-week experiment, and questions to ask
Useful AI work for teachers starts with clear learning goals, verified content, and careful boundaries.
Skills to build
- Domain knowledge: know the curriculum, standards, student development, classroom norms, and assessment purpose.
- Verification: check facts, examples, answer keys, sources, and alignment before use.
- Communication: explain AI use clearly when policy requires it and write family-facing language with care.
- Judgment: decide when an AI draft supports learning and when it weakens student agency or assessment validity.
- Privacy and security: recognize student records, identifiable work, grades, behavior notes, and disability information.
- Workflow thinking: map how a draft becomes a lesson, assessment, record, family message, or professional decision.
- Accessibility: review reading order, captions, alternatives, contrast, headings, plain language, and assistive-technology needs.
Playbook
First-week experiment: improve one ungraded practice activity
Goal: Create a teacher-reviewed practice activity faster without student data. Preparation: Use an approved tool, a public learning objective, trusted sources, and your own notes. Steps: ask for draft questions, verify every answer, revise for age level, check accessibility, remove weak questions, and compare with a version you would normally write. Success measures: accurate questions, clearer wording, better misconception support, less formatting time, and no privacy concern. Stop conditions: the activity becomes graded or high-stakes, the tool invents facts, the content is biased or inappropriate, or student data would be needed. Reflection: What helped? What required teacher judgment? Did the activity support the learning goal? What should stay human-only?
- Use no student records.
- Keep a review note showing what the AI got wrong.
- Ask a colleague to review one item for clarity.
- Do not expand to grading or accommodations.
Questions to ask your employer or institution
- Which AI tools are approved for teacher planning, communication, translation, and assessment support?
- What student information may never be entered into AI tools?
- What disclosure is required to students, families, colleagues, or administrators?
- How should AI-assisted materials be stored, labeled, and retained?
- What copyright and licensing rules apply to prompts, source materials, images, and outputs?
- Who reviews AI-assisted rubrics, assessments, translations, and family communications?
- What accessibility checks are required before materials are shared?
- Who is accountable if AI-assisted content is inaccurate, biased, inaccessible, or misused?
Avoidable errors
Common mistakes and better approaches
Using AI to grade or decide consequences.
Better approach: Use AI only for preparation or feedback-language drafts; keep grading and consequences with authorized educators.
Pasting student records into an unapproved tool.
Better approach: Use approved systems and minimize data; avoid identifiable student information unless policy explicitly allows it.
Trusting an answer key because it looks polished.
Better approach: Verify every answer against trusted sources and teacher knowledge.
Using AI materials without accessibility review.
Better approach: Check reading order, captions, alternatives, headings, links, and plain language.
Letting AI replace student thinking.
Better approach: Design tasks that require students to explain, revise, discuss, or apply ideas.
Remember this
Key takeaways
- AI may help with drafts, not teacher judgment.
- Do not upload student records into unapproved tools.
- Lesson drafts need accuracy, bias, accessibility, age, source, and curriculum checks.
- Rubrics and assessments need teacher and institutional review.
- Feedback preparation is not the same as grading.
- Student agency and academic integrity should shape AI use.
- Institutional policy and local law matter.
Questions readers ask
Frequently asked questions
Can AI grade student work?
Do not delegate grading decisions to AI. AI may help draft feedback stems or organize rubric language, but an authorized educator should evaluate student work under institutional policy.
Can I use AI to write lesson plans?
AI can draft lesson ideas, examples, and activities. A teacher should check curriculum alignment, factual accuracy, learning goals, age appropriateness, accessibility, bias, and classroom fit.
Can AI help with translation?
AI may help draft simple, non-sensitive translations, but high-stakes communication needs qualified review and institutional process. Do not use unreviewed AI translation for discipline, disability services, health, legal rights, or safeguarding messages.
Do I need to disclose AI use?
Disclosure depends on institutional policy, assignment rules, and jurisdiction. Ask your school or district how to label AI-assisted materials and student-facing guidance.
Can students use AI?
That depends on the learning objective and policy. Clear instructions should say when AI is allowed, what help is allowed, what must be cited or disclosed, and what work must show the student’s own thinking.
Sources and review notes
Sources were accessed on the dates shown. Links open the original organization’s page.
- SRC-01Secondary School Teachers, Except Special and Career/Technical Education (25-2031.00)U.S. Department of Labor, O*NET OnLine · Accessed 2026-06-20
- SRC-02High School Teachers: Occupational Outlook HandbookU.S. Bureau of Labor Statistics · Published 2025-08-28 · Accessed 2026-06-20
- SRC-09Generative AI and Jobs: A global analysis of potential effects on job quantity and qualityInternational Labour Organization · Published 2023-08-21 · Accessed 2026-06-20
- SRC-10AI and workOrganisation for Economic Co-operation and Development · Accessed 2026-06-20
- SRC-12Artificial Intelligence Risk Management Framework (AI RMF 1.0)National Institute of Standards and Technology · Published 2023-01-26 · Accessed 2026-06-20
- SRC-13Artificial Intelligence Risk Management Framework: Generative Artificial Intelligence ProfileNational Institute of Standards and Technology · Published 2024-07-26 · Accessed 2026-06-20
- SRC-14Guidance for generative AI in education and researchUNESCO · Published 2023-09-07 · Accessed 2026-06-20
- SRC-15FERPAU.S. Department of Education, Student Privacy Policy Office · Accessed 2026-06-20
- SRC-16Web Content Accessibility Guidelines (WCAG) 2.2World Wide Web Consortium · Published 2023-10-05 · Accessed 2026-06-20
- SRC-25Copyright and Artificial IntelligenceU.S. Copyright Office · Accessed 2026-06-20