How AI Helps Co-Teachers Plan for Both General and Special Education
Co-teaching — a general education teacher and a special education teacher sharing responsibility for instruction in the same classroom — is the most common service delivery model for inclusive education. Approximately 40% of students with disabilities spend 80% or more of their school day in general education classrooms (NCES, 2022), and in most of those classrooms, the special education teacher provides support through some form of co-teaching arrangement.
The research on co-teaching is frustratingly mixed. When done well — shared planning, shared instruction, shared assessment — co-teaching improves outcomes for students with disabilities without negatively affecting general education students (Murawski & Swanson, 2001). When done poorly — which is far more common — it devolves into "one teach, one hover," where the special education teacher circulates the room helping individual students while the general education teacher does all the actual teaching. In surveys, special education co-teachers report feeling like "glorified paraprofessionals" in more than half of co-teaching partnerships (Scruggs, Mastropieri & McDuffie, 2007).
The single greatest barrier to effective co-teaching is planning time. Co-teachers need to plan together — which model to use, who does what, how to differentiate materials, how to embed IEP accommodations — but shared planning periods are rare. One national survey found that 63% of co-teaching pairs had zero common planning time built into their schedules (Conderman, 2011). AI can compress planning time dramatically, generating differentiated materials, role-division plans, and accommodation-embedded lessons from a single prompt.
The Six Co-Teaching Models
| Model | How It Works | Best For | AI Planning Support |
|---|---|---|---|
| One Teach, One Observe | One teacher instructs; the other collects specific data on student behavior, engagement, or academic performance | Gathering baseline data; monitoring IEP progress; identifying struggling students | AI generates observation checklists, data collection forms, specific behaviors to watch for |
| One Teach, One Assist | One teacher instructs; the other circulates to provide individual support | Whole-class instruction where some students need proximity support or redirection | AI generates "assist" scripts with specific interventions for struggling students, based on IEP goals |
| Station Teaching | Teachers divide content into segments; each teaches a station while students rotate. A third station may be independent work. | Practice and application lessons; review lessons; content that can logically split into parts | AI generates complete station plans including materials, timing, and independent station activities |
| Parallel Teaching | Class is split in half; each teacher teaches the same content to a smaller group | Reducing student-to-teacher ratio; content that benefits from more participation opportunities | AI generates the same lesson in two versions — one standard, one with embedded accommodations |
| Alternative Teaching | One teacher works with the large group; the other pulls a small group for pre-teaching, re-teaching, or enrichment | Pre-teaching vocabulary before a lesson; re-teaching after assessment; enrichment and extension | AI generates pre-teach and re-teach mini-lessons aligned to the main lesson content |
| Team Teaching | Both teachers instruct simultaneously — co-presenting, debating, or taking complementary roles | Modeling discussion; content with multiple perspectives; when both teachers have strong content knowledge | AI generates lesson scripts with both teacher roles marked and transitions planned |
Key insight from the research: Effective co-teaching pairs use multiple models, not just one. Friend (2008) found that the most successful co-teaching partnerships rotated through 3-4 models per week, matching the model to the lesson objective. Defaulting to "One Teach, One Assist" every day is the most common failure mode.
AI Prompts for Co-Teaching Planning
Co-Taught Lesson Plan Generator
Generate a CO-TAUGHT LESSON PLAN for a Grade [X] [subject]
class on [topic/standard].
CLASS COMPOSITION:
- Total students: [number]
- Students with IEPs: [number]
- IEP accommodations that must be embedded:
[List specific accommodations, e.g.:]
- Student A: Extended time, reduced problem sets, text
read aloud
- Student B: Preferential seating, check for understanding
every 5 minutes, visual schedule
- Student C: Sentence frames for all written responses,
word bank, graphic organizer
CO-TEACHING MODEL FOR THIS LESSON: [select from the six
models, or say "recommend the best model"]
LESSON PLAN FORMAT:
1. LEARNING OBJECTIVE: What all students will know/do by
the end (same objective for all students)
2. ROLE DIVISION TABLE:
| Time | General Ed Teacher Does | Special Ed Teacher Does |
|:-----|:----------------------|:----------------------|
| 0-5 min | [specific action] | [specific action] |
| 5-15 min | [specific action] | [specific action] |
| 15-30 min | [specific action] | [specific action] |
| 30-40 min | [specific action] | [specific action] |
| 40-45 min | [specific action] | [specific action] |
3. MATERIALS — GENERAL VERSION: The standard-level materials
for the lesson (worksheet, reading, activity instructions)
4. MATERIALS — ACCOMMODATED VERSION: The same materials with
IEP accommodations embedded. Mark each accommodation so
the special ed teacher can see what was modified and why.
Use [brackets] to indicate modifications:
[REDUCED: 8 problems instead of 15]
[READ ALOUD: This section should be read to the student]
[SENTENCE FRAME: Response template provided]
5. FORMATIVE ASSESSMENT: A quick check (exit ticket or
observation) that both teachers can use. Include
accommodated version.
6. DATA COLLECTION NOTE: What the special ed teacher should
document for IEP progress monitoring during this lesson.
Station Teaching Materials Generator
Generate STATION TEACHING MATERIALS for a co-taught Grade [X]
[subject] lesson on [topic].
CLASS SETUP: 3 stations, 15 minutes each, students rotate.
STATION 1 — TEACHER-LED (General Ed Teacher):
- Direct instruction or guided practice on [specific skill]
- Include a mini-lesson script (5 min instruction + 10 min
guided practice)
- Materials: [worksheet/manipulatives/whiteboard activity]
STATION 2 — TEACHER-LED (Special Ed Teacher):
- Targeted instruction on the SAME skill but with:
- More scaffolding
- Slower pacing
- Embedded IEP accommodations for: [list students and
accommodations]
- Smaller group = more individual response opportunities
- Materials: Accommodated version of Station 1 materials
STATION 3 — INDEPENDENT/COLLABORATIVE:
- Practice activity students can complete WITHOUT a teacher
- Must be self-checking or peer-checkable
- Must be engaging enough to maintain on-task behavior for
15 minutes without teacher supervision
- Include clear written directions (assume no teacher nearby
to explain)
- Include a "finished early" extension activity
LOGISTICS:
- Rotation signal: [timer, bell, verbal cue]
- Student grouping recommendation (which students to which
stations and in what order — students who need the most
support should go to Station 2 first while content is
fresh)
- Transition procedure: 1-minute transition between stations
IEP Accommodation Embedding
I have the following STANDARD LESSON PLAN for Grade [X]
[subject]:
[Paste the existing lesson plan]
The following students have IEP accommodations that must be
embedded into this lesson:
Student 1: [Name/initial]
- Accommodation: [e.g., "Extended time on all assignments"]
- Accommodation: [e.g., "Text read aloud"]
- Accommodation: [e.g., "Reduced assignment length — no more
than 10 problems"]
Student 2: [Name/initial]
- Accommodation: [e.g., "Graphic organizer for all writing"]
- Accommodation: [e.g., "Preferential seating near the board"]
- Accommodation: [e.g., "Check for understanding every 5-10
minutes"]
Student 3: [Name/initial]
- Accommodation: [e.g., "Manipulatives available for all
math computation"]
- Accommodation: [e.g., "Word bank for vocabulary-heavy
tasks"]
GENERATE:
1. An ACCOMMODATION INTEGRATION TABLE showing exactly where
in the lesson each accommodation applies:
| Accommodation | Lesson Phase | How to Implement |
|:-------------|:-------------|:-----------------|
2. MODIFIED MATERIALS for each student who needs them,
clearly marked with what was changed and why.
3. A QUICK-REFERENCE CARD the special ed teacher can keep
on a clipboard during the lesson — one line per student,
listing their accommodations and what to watch for.
Planning Without Common Planning Time
The reality: most co-teaching pairs don't have shared planning time. AI can create a planning workflow that minimizes the need for synchronous meetings.
Generate a CO-TEACHING PLANNING TEMPLATE that two teachers
can complete ASYNCHRONOUSLY (without meeting in person).
FORMAT: A shared document (Google Doc or equivalent) with
the following sections:
PART 1 — GENERAL ED TEACHER COMPLETES (5-10 minutes):
- Topic/standard for the week: ___
- Planned activities: ___
- Materials already created: ___
- Assessment plan: ___
- "I need help with...": ___
PART 2 — SPECIAL ED TEACHER COMPLETES (5-10 minutes):
- IEP goals that can be addressed this week: ___
- Students who need pre-teaching on this topic: ___
- Accommodation modifications needed: ___
- Suggested co-teaching model for each day: ___
- "I can do...": ___
PART 3 — AI-GENERATED (they paste both parts into AI):
Given the completed Parts 1 and 2, generate:
- Daily co-teaching model recommendation (Mon-Fri)
- Role-division table for each day
- Accommodation-embedded materials
- Data collection opportunities for IEP goals
This creates a planning workflow that takes each teacher
10 minutes individually + 5 minutes reviewing the
AI-generated plan = 25 total minutes instead of a
45-minute co-planning meeting.
Data Collection for IEP Progress Monitoring
Generate IEP PROGRESS MONITORING TOOLS that the special
education teacher can use DURING co-taught lessons without
disrupting instruction.
STUDENT: [Name/initial]
IEP GOAL: [Paste the exact IEP goal]
GENERATE:
1. A TALLY SHEET for recording goal-related behaviors during
class. Format:
Date: ___ | Lesson: ___
Opportunities observed: ___ (how many times the student
had a chance to demonstrate the skill)
Successful demonstrations: ___
Percentage: ___/___= ___%
2. A 1-MINUTE PROBE the teacher can administer individually
during independent work time (while the general ed teacher
monitors the class). Must be quick, quiet, and
curriculum-relevant.
3. A WEEKLY SUMMARY TEMPLATE:
Week of: ___
Goal: ___
Data points this week: ___
Trend: Increasing / Stable / Decreasing
Action: Continue current intervention / Modify / Intensify
Notes for the team: ___
4. GRAPHING TEMPLATE: A simple chart where the teacher can
plot weekly percentages. Include the IEP goal line
(the target percentage from the IEP) and space for
20 weeks of data.
Avoiding "One Teach, One Hover"
The default co-teaching failure mode needs to be actively resisted. Here's what it looks like and how to prevent it:
| "One Teach, One Hover" Sign | What to Do Instead | AI Can Generate |
|---|---|---|
| Special ed teacher only helps individual students | Special ed teacher leads instruction for part of the lesson | Station teaching or parallel teaching lesson plans where both teachers have instructional roles |
| Same co-teaching model every day | Rotate through 3-4 models weekly | A weekly model rotation plan matched to lessons |
| Only the general ed teacher makes instructional decisions | Both teachers contribute to planning | Asynchronous planning templates with shared decision-making |
| Special ed teacher doesn't know the lesson plan until it starts | Both teachers review the plan before class | A plan-review checklist that takes 3 minutes and ensures both teachers know their role |
| Students see the special ed teacher as "the helper" | Special ed teacher instructs the whole class sometimes | Lessons where the special ed teacher leads instruction while the general ed teacher supports |
Key Takeaways
- Co-teaching only works when both teachers teach. The research is clear: "one teach, one hover" is the default failure mode, and it produces no better outcomes than a single teacher alone (Scruggs et al., 2007). AI-generated role-division tables make both teachers' responsibilities explicit before the lesson starts.
- Six co-teaching models exist — use at least three each week. Friend (2008) found that effective co-teaching pairs rotate models based on lesson objectives. AI can recommend the best model for each day's lesson and generate materials matched to that model.
- IEP accommodations must be built INTO materials, not added as afterthoughts. When accommodations are not embedded in the lesson materials, they get forgotten during instruction. EduGenius can generate standard and accommodated versions of the same lesson simultaneously, saving significant co-planning time.
- Common planning time is ideal but not required. Asynchronous planning workflows — where each teacher completes their part independently and AI generates the integrated plan — can reduce the need for synchronous meetings while maintaining co-teaching quality.
- Data collection happens during instruction, not after. The special education teacher's unique contribution during co-taught lessons includes monitoring IEP goals in real-time. AI-generated tally sheets and quick probes make this practical without disrupting the lesson flow.
See How AI Makes Differentiated Instruction Possible for Every Teacher for differentiation frameworks. See Accessibility in AI Education — Making Content Work for All Students for universal design. See AI for Differentiated Math Instruction — From Concrete to Abstract for CRA-based math differentiation. See Creating Student-Centered Learning Menus with AI for choice-based differentiation within co-taught classrooms.
Frequently Asked Questions
What if the co-teachers have different teaching philosophies?
This is common and can actually be productive. Station teaching and parallel teaching models allow each teacher to use their preferred approach with their group. Team teaching, where both co-present, requires more philosophical alignment. Start with station teaching (where each teacher has autonomy over their station) and move toward team teaching as the relationship develops. AI-generated materials provide a common foundation that both teachers can adapt to their style.
How do I ensure the special education teacher isn't just seen as "the helper"?
Two strategies: First, have the special education teacher lead whole-class instruction regularly — especially for content they have expertise in (study skills, test-taking strategies, metacognitive strategies). Second, during parallel teaching, give the special education teacher the "general" group sometimes while the general education teacher takes the smaller group. When students see both teachers in both roles, the "helper" perception fades.
What about when the special education teacher serves multiple classrooms?
This is the most common and most challenging scenario. A special education teacher who co-teaches in 3-4 different classrooms can't deeply co-plan with 3-4 different general education teachers. The asynchronous planning template becomes essential — 10 minutes per class, AI-generated integration plan, 3-minute review before class. Prioritize models that don't require extensive co-planning: station teaching (where the special education teacher runs one predictable station format across all classes) is more sustainable than team teaching.
Can AI replace co-planning meetings entirely?
No. AI can reduce the frequency and length of co-planning meetings, but it can't replace the relationship-building and real-time problem-solving that happens face to face. The research consistently shows that the quality of the co-teaching relationship predicts the quality of co-taught instruction (Gately & Gately, 2001). Use AI to handle materials generation and logistics, then use whatever face-to-face time you have for student-specific discussions, relationship maintenance, and instructional decision-making.
Next Steps
- How AI Makes Differentiated Instruction Possible for Every Teacher
- Accessibility in AI Education — Making Content Work for All Students
- AI for Differentiated Math Instruction — From Concrete to Abstract
- Creating Student-Centered Learning Menus with AI
- Using AI to Support Students with Speech and Language Delays
- AI for Mathematics Education — From Arithmetic to Algebra