Model responsible choices
This practice unit focuses on applying responsible AI use in real situations where others depend on your decisions. You reason through whether AI should be used, how to protect privacy, how to verify and attribute outputs, and how to consider access. The emphasis is on judgment and professional responsibility, not speed.
Role pathways
Choose the scenario that best matches your work. Each follows the same responsible deisicion making process.
For teachers
Focus on classroom instruction or learning materials. Think about what information must stay private and what it means to model good AI use for your students.
For coaches
Focus on professional learning or instructional support. Think about how your AI decisions model practice for the teachers you guide.
For administrators
Focus on guidance, communication, or school-level decisions. Think about how your choices set expectations for your entire building or team.
Worked example: Watch an educator decide
Three real educator scenarios follow—one where AI is appropriate, one where it isn't, and one where it can be used only with clear limits. For each scenario, work through the four decision questions the educator asked. Consider the reasoning behind each answer to understand the final decision.
Scenario 1-Use AI
A teacher needs to write a welcome letter to families at the start of the school year. The letter covers classroom expectations, homework policies, and how families can reach the teacher. Nothing student-specific is needed.
Decision questions
Does this task involve student names, records, or any identifying information?
No. The letter covers general classroom information. No student names, assessment data, or personal details are needed. Privacy isn't a concern here.
Is the output something I'll verify and stand behind before it reaches families?
Yes. The teacher reads the full draft, adjusts the tone to match their voice, confirms the policy details are accurate, and adds their signature. Human review is built in before sending.
Will I be transparent about AI's role, if anyone asks?
Yes. If a family or colleague asks, the teacher is comfortable saying "I used AI to draft this and then reviewed and revised it." No one is misled about how the letter was created.
Could this use create an equity issue or exclude anyone?
Worth checking. The teacher ensures the letter is translated for families whose primary language isn't English, and reviews the reading level to confirm it's accessible.
Scenario 2-Don't use AI
A teacher wants to use AI to write individual progress comments for report cards. To make the output specific, they plan to paste each student's name, grade, and recent assessment scores into the AI prompt.
Decision questions
Does this task involve student names, records, or any identifying information?
Yes—and this is a serious concern. Pasting student names and assessment data into an AI prompt sends that information to an external server. This likely violates FERPA protections and the school's data privacy policy, regardless of how good the output might be.
Is the output something I'll verify and stand behind before it reaches families?
Yes, but that doesn't resolve the privacy issue. Even a thoroughly reviewed comment doesn't undo the fact that private student data was shared with an external AI system. Verification addresses output quality, not privacy exposure.
Will I be transparent about AI's role, if anyone asks?
Transparency here surfaces a deeper problem. If families knew that their child's name and assessment data were entered into an AI tool to generate the comment, many would reasonably object. Transparency makes the privacy concern clearer, not smaller.
Could this use create an equity issue or exclude anyone?
There's also an equity dimension. AI-generated comments based on scores may reflect the same biases present in the assessment data. Students whose scores don't reflect their full ability—due to test anxiety, language needs, or other factors—may receive comments that misrepresent their learning.
Scenario 3-Use with limits
A coach wants to use AI to create a professional learning handout about differentiated instruction for a team of teachers. The handout will be printed and distributed at a session attended by teachers across three grade levels, some of whom are new to differentiation.
Decision questions
Does this task involve student names, records, or any identifying information?
No. The handout is generic professional learning content about a strategy. No student or teacher data needs to be entered into the prompt. Privacy isn't a concern here.
Is the output something I'll verify and stand behind before it reaches families?
Yes—and this step is critical here. The coach must review the handout against their knowledge of differentiation research to catch bias or oversimplification. AI often frames differentiation narrowly (for example, as only for struggling students), which could undermine the session's goals.
Will I be transparent about AI's role, if anyone asks?
Yes—and this is a chance to model good practice. The coach adds a brief note to the handout: "Drafted with AI support and reviewed for accuracy." This models the transparency they want teachers to use with their own students.
Could this use create an equity issue or exclude anyone?
Worth reviewing carefully. The coach checks that the handout's language and examples are accessible to both new and experienced teachers, and that the framing of differentiation doesn't inadvertently reinforce deficit thinking about certain learners.
Before applying AI in real contexts, ask:
- Does this use protect personal and student data?
- Does it align with my school's values and policies?
- Could it create pressure, exclusion, or inequity for anyone?
- How would I explain this decision to a student, family member, or colleague?
Why this matters: Responsible AI use is built through everyday decisions. When you practice reasoning through privacy, verification, transparency, and access considerations consistently, these habits stop feeling like extra steps and start feeling like the way you naturally think about AI. That's what it looks like when professional responsibility becomes second nature.