The problem

AI training usually misses the people doing the hardest paperwork

  • Most sessions teach the tool, not the workday — nothing changes on Sunday night when the present levels are still blank.
  • Almost nothing addresses students with disabilities: nonverbal students, AAC users, students who cannot follow a verbal command.
  • Privacy guidance is either absent or so vague that staff quietly paste student data into a chat box anyway.
  • Prompt lists get shared once and then lost, because no one wrote down the role, scope, or rules behind them.
Concrete outcomes

What you can do differently afterward

  • Draft a present-levels statement from your own raw data notes in under 10 minutes, then edit it as the professional — instead of staring at a blank template on a Sunday night.
  • Turn a week of behavior tally sheets into a parent-ready summary that names patterns, antecedents, and next steps in language a family can actually use at home.
  • Run a de-identification pass on any text before it touches an AI tool, using a rule you can state out loud to your principal.
  • Stand up three named assistants — documentation, behavior, and communication — each with a written role, boundaries, and an activation prompt you reuse all year.
  • Rewrite a safety or lockdown procedure for a student who cannot follow verbal commands, and defend every change in an IEP or safety-team meeting.

Training modules

AI foundations for special educators

What these tools actually do, where they fail, and how to tell the difference — in plain language, with no assumption of a tech background.

Prompting for IEP and documentation work

Reusable prompt chains for present levels, goal drafting, and parent-friendly behavior summaries. You keep the professional judgment; AI handles the first draft.

Student data and privacy discipline

De-identification habits, what never goes into a prompt, and how to build a workflow your district can review without wincing.

Building your own AI team

The Command Center approach, scaled down: three specialized assistants instead of twenty-four, set up in an afternoon.

Non-negotiables

What “responsible” means here

  • AI drafts. Educators decide. Nothing goes to a family unreviewed.
  • No identifiable student data in general-purpose AI tools.
  • Every prompt system is documented so a colleague can reuse it.
  • If a tool does not save real time in a real teaching day, it does not make the training.

Why Karey

Still in the classroom

Currently practicing special education teacher — 23+ years, Rochester City School District.

Google Certified Educator

Level 1 and Level 2 certified.

AI InnoVision certified

AI Consultant certification through Alicia Lyttle's AI InnoVision program.

Former EMT, ICS-trained

12 years in emergency medical services; Incident Command System trained.

Many AI trainings start with the technology. Starr TeachForward starts with the educator’s actual workday. Karey is in a special education classroom this week. Every prompt, workflow, and privacy rule in this training has to survive a real teaching day before it gets taught.

The next step after the free resources

Build Your AI Team in 5 Days

Five days, one assistant at a time. You finish with a small, named AI team built on the same role-and-scope structure behind the Command Center — documentation, behavior, communication — plus the privacy rules to use it with real caseloads.

  • Day 1: Your documentation assistant and its written role boundaries
  • Day 2: Behavior data in, parent-ready summary out
  • Day 3: De-identification workflow you can defend to an administrator
  • Day 4: Communication assistant for families and service providers
  • Day 5: Wiring the team together and reusing it all year
$197one-time

Self-paced, lifetime access. Built and taught by a currently practicing special education teacher.

Ask about team or district pricing

Free starting points