What happened
The response, method, time, help requested, and lesson context remain visible.
A more connected way to learn is taking root.Join early access
How Curiosity Tree works
A clear path, with help that knows where to begin.
The curriculum stays coherent. The presentation adapts. Evidence—not an opaque model—determines when a learner needs support and when a lesson is ready to turn golden.
Current Knowledge TreeFractions & Decimals8 of 24 golden
FoundationsMastered
Compare fractionsCurrent lesson
Support pathCommon denominatorsReturn point saved
Add unlike fractionsNext
The learning loop
Structured enough to be dependable. Adaptive enough to be personal.
Curiosity Tree is designed to respond to what the learner demonstrates without handing curriculum control to a black box.
Evidence with context
Practice is for learning—not permanent punishment.
What it demonstrated
Independent success counts differently from an answer completed after revealing help.
What is ready
Several qualifying pieces of evidence—not one lucky answer—support proficiency.
What still lasts
Later retrieval and transfer confirm that understanding remained available.
AI with boundaries
The model may advise. The learning system decides.
Large language models (LLMs) are AI systems trained to recognize patterns in language and generate conversational responses. They can make tutoring more responsive by offering alternate explanations, hints, practice variations, and careful interpretation of open responses. In Curiosity Tree, however, they serve as bounded teaching tools: they cannot unlock lessons, assign mastery, calculate grades, or quietly rearrange the curriculum.
- Deterministic checks wherever possible
- Structured, source-aware model responses
- Parent-visible evidence and overrides
- Human review for high-stakes judgments
See where thoughtful adaptation can lead.
Join the early-access community for prototype invitations and future family pilot news.
Join early access