Education + child-centered AI governance · Independent · research, science, design, and AI
AI Use Decision Lab
Live local-first decision-support product · v2 August 2026
Help families, educators, schools, clinicians, and product teams evaluate one specific child-facing AI use before adoption.
Controlled visual demonstration
Current live interface
This sandboxed preview shows the current public interface. Open the full site to use the build in its intended window.
What I noticed
Families and schools are being asked whether children should use AI as though every tool, task, age, data practice, and level of adult oversight creates the same problem. The practical need is a transparent way to examine the role AI will play, what it may replace, what information it receives, and who remains accountable.
What I researched
The decision model draws on the American Academy of Pediatrics’ 2026 state-of-the-art review, UNICEF’s child-centered AI guidance, UNESCO’s generative-AI guidance for education, U.S. Department of Education FERPA guidance, the FTC’s amended COPPA Rule, and NIST’s Generative AI Profile. It avoids claiming that longitudinal neurological effects are already settled.
What exists
A five-stage, age-aware decision lab covering context, AI role, cognitive substitution, adult involvement, consequence level, data and human safeguards, inspectable decision logic, developmental capacities to protect, vendor or school questions, monitoring signals, dated source links, browser persistence, Markdown export, and print/PDF output.
Who it serves
Parents, caregivers, educators, schools, clinicians, youth-serving organizations, and teams designing or procuring child-facing AI.
What it demonstrates
Developmental-science translation, child-centered AI governance, privacy-aware product design, deterministic decision logic, evidence architecture, risk communication, and local-first implementation.
Current evidence
The canonical production release was browser-verified with two opposing scenarios: an unsupervised emotional-companion use for a young child was stopped with explicit reasons, while a bounded educational use for an older teen with complete safeguards advanced as a monitored trial. The tool does not diagnose children, certify products, determine legal compliance, or claim settled long-term neurological effects.
Verified build record
What is actually running, and what it is not.
LiveLive · v2, August 2026
The problem. Families and schools are asked whether children should use AI as though the tool, the task, the child's age, the data practices and the level of adult oversight all raised the same question. They do not, and the general answer is useless for the specific decision in front of someone.
What it does. Evaluates one specific child-facing AI use across five age-aware stages: context, the role AI will play, what cognitive work it may substitute for, adult involvement, consequence level, and data and human safeguards. The decision logic is inspectable rather than hidden, and the output names the developmental capacities to protect, the questions to put to a vendor or school, and the signals to monitor.
Built on. A static browser application with local persistence and dated source links, served from Vercel. The decision model draws on the AAP 2026 state-of-the-art review, UNICEF child-centred AI guidance, UNESCO guidance for education, US Department of Education FERPA guidance, the FTC's amended COPPA Rule, and the NIST Generative AI Profile.
What it does not establish. It does not diagnose children, certify products, determine legal compliance, or claim that long-term neurological effects are settled.
This record is generated from the same inventory as the complete build index: the production URL was checked at the last regeneration, and the status line above repeats what the build reports about itself rather than restating an intention.