Education-First
The EDU-FIRST guardrail wraps every LLM call, constraining output to Bloom Taxonomy and age-appropriate cognitive load.
The EDU-FIRST guardrail wraps every LLM call, constraining output to Bloom Taxonomy and age-appropriate cognitive load.
Dual defense: lexical rules + LLM Judge. COPPA, GDPR-aligned and Thai PDPA-aware. Data minimization by default.
All 13 microservices produce structured audit logs retained for 7 years. Every pedagogical decision is reproducible.
The Curriculum Engine automatically adapts complexity, strategy and tone to each age bucket.
Three interfaces, three journeys, one shared AIT multi-agent backend.
A multimodal Avatar driven by facial BlendShapes, voice and posture — bringing warmth and encouragement to every turn.
The EDU-FIRST model sits before and after every LLM call to validate pedagogy, age and Bloom level.
Regex + NLP lexical layer first, LLM Judge second. Uncertain cases always escalate to a human teacher.
Socratic / SocraticPlus / KnowledgeTutor / Wellbeing collaborate and route to the best model.
“AIT is the first pedagogical system that I trust to put AGI-grade reasoning at the service of every child — without sacrificing safety or auditability.”
“In SE Asia we have dreamed of equity in education for 30 years. AIT finally delivers a deployable, locally relevant AI teacher.”
“The cognitive-architecture behind AIT's Socratic routing is years ahead of the generic LLM wrappers we have seen in K12.”