Lesson 1173 of 1550
Using AI to redesign formative assessments
Use AI to redesign formative assessments so they reveal misconceptions, not just right or wrong answers.
Lesson map
What this lesson covers
Learning path
The main moves in order
- 1The premise
- 2Using AI to build a differentiated task bank
- 3The premise
- 4Using AI to generate rubric norming cases
Concept cluster
Terms to connect while reading
Section 1
The premise
Formative assessments often only tell you who got it wrong. AI can redesign items so wrong answers reveal which misconception the student holds.
What AI does well here
- Generate distractors tied to specific misconceptions.
- Tag each answer choice with the misconception it implies.
- Suggest a follow-up question per misconception.
What AI cannot do
- Know your specific class's history.
- Replace teacher observation.
- Decide what reteaching to plan.
Key terms in this lesson
Section 2
Using AI to build a differentiated task bank
Section 3
The premise
Differentiation is exhausting because each task takes time to write. AI can produce a tiered task bank for one goal in minutes.
What AI does well here
- Produce three tiers of tasks for one learning goal.
- Vary scaffolding, not the goal.
- Suggest a quick check that identifies which tier a student needs.
What AI cannot do
- Know your students by name.
- Account for IEP accommodations.
- Replace teacher judgment about who is ready for what.
Section 4
Using AI to generate rubric norming cases
Section 5
The premise
Norming sessions stall when teams cannot agree on examples. AI can generate sample student work calibrated to each rubric level for the conversation.
What AI does well here
- Generate work samples calibrated to each rubric level.
- Include borderline cases on purpose.
- Annotate which rubric criteria each sample meets.
What AI cannot do
- Replace real student work over time.
- Know your team's history of disagreement.
- Decide where the team should land.
Section 6
Using AI to do a tone pass on parent updates
Section 7
The premise
Parent updates often read as alarmed or defensive without meaning to. AI can do a focused tone pass that preserves every fact and ask.
What AI does well here
- Identify defensive or alarmed phrasing.
- Suggest neutral and warm rewrites.
- Preserve all facts and asks.
What AI cannot do
- Know the family's history.
- Decide what to escalate to a phone call.
- Predict the parent's reaction.
Section 8
Using AI to design PD from teacher survey results
Section 9
The premise
PD sessions miss when planners pick a topic without looking at survey results. AI can design a 60-minute session targeted at the top survey themes.
What AI does well here
- Cluster survey themes.
- Design a 60-minute session with active practice and a takeaway artifact.
- Suggest a follow-up step at 30 days.
What AI cannot do
- Know your faculty's relational dynamics.
- Replace facilitation skill in the room.
- Guarantee adoption.
Section 10
Using AI to check curriculum vertical alignment
Section 11
The premise
Vertical alignment work usually drowns in standards documents. AI can read across grade-level scope and sequences and flag gaps and unintended repetition.
What AI does well here
- Compare scope and sequence across grades.
- Flag concepts taught at the same depth twice.
- Identify gaps where a prerequisite is assumed but not taught.
What AI cannot do
- Know your district's adopted curriculum quirks.
- Replace a teacher articulation meeting.
- Resolve disagreements between grade-level teams.
Section 12
Using AI to troubleshoot a classroom routine
Section 13
The premise
Routines stall at one or two specific moments. AI can interview you about the routine and propose targeted fixes instead of overhauling everything.
What AI does well here
- Ask diagnostic questions about timing and signals.
- Propose two targeted fixes per stall point.
- Suggest a 5-day trial.
What AI cannot do
- See your room.
- Know your student rapport.
- Replace classroom management practice.
Section 14
Using AI to build a grading feedback bank
Section 15
The premise
Specific feedback is what students learn from, but writing it for every paper burns out teachers. AI can build a feedback bank tuned to the rubric.
What AI does well here
- Generate 15 to 20 short feedback snippets per rubric criterion.
- Vary tone for high, medium, and low performers.
- Tag each snippet with the misconception or strength it addresses.
What AI cannot do
- Know which student needs which snippet.
- Replace handwritten notes for high-stakes work.
- Account for student relationships.
Section 16
Using AI to draft a substitute plan from a unit map
Section 17
The premise
Sub plans default to busywork because writing real ones takes too long. AI can produce an executable plan a stranger can run, tied to the unit.
What AI does well here
- Produce a minute-by-minute plan a stranger can follow.
- Include classroom procedures and signals.
- Generate a one-paragraph debrief for the teacher's return.
What AI cannot do
- Know which sub will actually show up.
- Account for student behavior triggers.
- Guarantee fidelity to the lesson.
Section 18
Using AI to generate SEL warmups
Section 19
The premise
SEL feels disconnected when warmups are random. AI can generate brief warmups tied to the week's academic content so SEL reinforces, not interrupts.
What AI does well here
- Generate 5 warmups per week, each under 5 minutes.
- Tie each to the week's academic theme.
- Vary modality (write, share, draw, move).
What AI cannot do
- Know specific student trauma history.
- Replace counselor or specialist input.
- Guarantee emotional safety.
End-of-lesson quiz
Check what stuck
15 questions · Score saves to your progress.
Tutor
Curious about “Using AI to redesign formative assessments”?
Ask anything about this lesson. I’ll answer using just what you’re reading — short, friendly, grounded.
Progress saved locally in this browser. Sign in to sync across devices.
Related lessons
Keep going
Adults & Professionals · 40 min
Formative Assessment Prompts: Quick Checks That Actually Inform
Exit tickets and quick checks are only useful if they surface what students actually don't understand. AI can generate targeted formative probes that reveal misconceptions, not just surface recall.
Adults & Professionals · 40 min
Differentiated Instruction Generators: One Lesson, Every Learner
Differentiation used to mean creating three separate versions of every handout. AI can generate tiered materials from a single prompt — if you describe the learner profiles clearly.
Adults & Professionals · 40 min
Grading Feedback Automation: Actionable Comments at Scale
Margin comments like 'good job' or 'needs work' don't help students improve. AI can generate specific, growth-oriented feedback comments aligned to rubric criteria — but teachers must decide the score and review every comment.