The AI-cheating crisis is the rare EdTech problem with national-press urgency, teacher-level pain, and no credible incumbent answer. Detection vendors are publicly conceding the arms race. Schools are reverting to pen and paper. Every ELA and humanities teacher in the country is improvising policy alone. The buyer is pre-sold on the problem; what doesn't exist yet is a product whose answer survives scrutiny.
The wedge user is the grade 7–12 writing teacher — old enough that students are sophisticated digital cheaters and over 13 (webcam tier available), young enough that take-home writing is the backbone of the pedagogy. She adopts tools herself when they're free and browser-based; her classroom runs on Chromebooks; she cannot get a new desktop app through district IT, and she doesn't have to — TeacherAware is a URL.
The under-13 profile (no webcam, screen + provenance only) extends the same foundation down into middle school without touching the COPPA line. District administration enters later, as the license buyer, after teachers already use it.
The strategic asymmetry: the provenance foundation costs nothing to run — the chain, the citation library, deterministic proctoring, and the dashboard are all local computation and cheap storage, no model calls. So the complete integrity answer is free forever. What costs real money — AI deep-passes, the screen-depth judge, the in-app constitutional collaborator — is precisely what schools will pay for once the free tier has proven the product in their own classrooms.
| Tier | Who buys | What lights up |
|---|---|---|
| Classroom (free) | Individual teacher, no approval needed | Workspace, chain + verification, citation library, deterministic proctoring, dashboard |
| School | Principal / department | AI deep-pass on flagged windows, screen depth (used? relevant?), constitutional writing collaborator |
| District | District IT + curriculum | Managed-Chromebook policy templates, multi-monitor awareness, stylometric backstop (roadmap), rostering/SSO (roadmap) |
The escalation-gate architecture is also the unit-economics story: AI cost scales with flags, not with hours of writing, so School-tier pricing can be flat per school without inference-cost anxiety.
| Incumbent | What they sell | Why TeacherAware wins the comparison |
|---|---|---|
| Turnitin | Post-hoc detection scores | Detection is the losing race, publicly conceded; a probability score is not evidence. Provenance is. |
| GoGuardian / Securly | Always-on device surveillance | Surveillance without provenance — it watches everything and can prove nothing about how a document was written. TeacherAware watches only sanctioned sessions and proves exactly that. |
| Lockdown browsers | Exam-mode lockouts | Built for tests, unusable for the weeks-long composition process where the cheating actually happens. |
| Google Classroom / Docs version history | Coarse revision snapshots | Version history is spoofable by typing over pasted text and records no integrity chain, no proctoring, no citation provenance. [There's something else that all of these share: they cost money. TeacherAware is FREE. Because it cost literally $5,000 to build and deploy at scale. $10-$20k of marketing spend is all it needs. That's how you grab market share. Then OpEx is three high level folks vs whatever grammarly has to spend. So this is kindof a new world for VCs: the startup costs are minimal, the OpEx and R&D is a fraction (Built around AI productivity from the start), so margin, when the auto-customizing for each school (after they do a wholesale dump of everything digital data ever (no names)) LMS is unveiled, which costs half as much as the competiton, so TAM is ~the LMS market-size value/2. Let me know if you agree. So in an ideal world, we're talking, i think, when all the math is done, 100x multiples for early investors, not 10x. ] |