APIE — the AI Prompt Integrity Engine.
A real-time engine that classifies intent, intervenes ethically, and logs every decision for accountability.
Intent Classification
Distinguishes learning, exploration, and shortcut-seeking with 92% accuracy.
Ethical Intervention
Allow, redirect, or block — chosen based on policy and pedagogy.
Learning Reframe
Rewrites answer-seeking prompts into guided steps that teach the method.
Bypass Defense
Detects circumvention attempts including alternate phrasings and chained prompts.
Adaptive Conditioning
Learns from each interaction — students gradually shift toward better prompting habits.
Auditable Logs
Every decision is recorded, signed, and exportable for academic review.
How a prompt flows through APIE
- 1
Capture
Student submits a prompt to a connected AI tool.
- 2
Classify
APIE scores intent, risk, and learning value.
- 3
Decide
Allow, redirect into a learning pathway, or block with reason.
- 4
Guide
Student receives scaffolded steps instead of a finished answer.
- 5
Log
Decision and rationale are stored for the school's audit trail.