Teacher-controlled AI
AI drafts and rubric suggestions stay behind teacher review, approval, override, and release decisions.
This page is built for school leaders, IT teams, and curriculum teams who need to separate current product proof from validation work and roadmap intent.
Similar education platforms sell activity breadth, AI prep, accommodations, and analytics. GoHiMark should compete by making those ideas inspectable: show the workflow, show the evidence, and show the boundary before a claim becomes a promise.
AI drafts and rubric suggestions stay behind teacher review, approval, override, and release decisions.
The public builder now maps to the Storybook question taxonomy and exposes question type, Bloom, grade, and validation detail.
ACARA and syllabus language is framed as an alignment workflow until descriptor-level coverage manifests are validated.
Course delivery, gradebook write-back, and reporting are shown as staged integration work with explicit proof gates.
Assessment and classroom platforms lead with many activity types.
GoHiMark should lead with taxonomy depth, validation controls, and teacher approval.
AI teacher tools focus on lesson prep, rubrics, reports, and question generation.
GoHiMark should show prep speed only when paired with review, provenance, and export readiness.
Mature school platforms make learner access and accommodations visible.
GoHiMark should expose accessibility checks per question type and per delivery mode.
LMS and assessment platforms sell insight, not raw charts.
GoHiMark should show evidence packs, intervention candidates, and confidence caveats.
Route buyers from proof center to builder, then from builder to a governed walkthrough request. The sequence is stronger than a generic demo CTA because it lets teachers and procurement inspect the product before a sales conversation.
These are the gates the marketing app now uses to keep public copy useful without letting it outrun the application.
Marketing routes should render the builder section and proof center in local and Docker runtime.
The builder must include every canonical question code from the Storybook question taxonomy.
Marketing copy must avoid unsupported full-coverage, auto-grading, outcome, and LMS write-back claims.
Security, privacy, AI routing, data retention, and hosting claims need a named evidence artifact.
Claims shown without development reveal
Implemented or proof-backed capabilities
Hidden until explicitly requested
Public claims are listed as evidence records. Search and filter without making development-only claims public by accident.
documentation-governance/generated/question-type-subject-coverage-2026-07-01/question-type-subject-coverage.md
documentation-governance/generated/unfounded-claims-2026-07-01/unfounded-claims-validation.md
documentation-governance/generated/marketing-bento-governance-2026-06-30/marketing-bento-validation.md
documentation-governance/generated/question-type-full-spectrum-audit-2026-07-01/question-type-full-spectrum-audit-parts-01-10.zip