Engage
Understand AI and ask better questions
Notice how AI appears in life, learning and media; distinguish questions, data, rules and results instead of treating output as fact.
AI literacy · Expression · STEM · Bilingual
Help kids ask better questions
and explain their ideas with evidence.
AIcoach4kid is not a chatbot that completes homework. It helps primary and secondary learners build the technical knowledge to understand AI, durable thinking and creative skills, and future-ready attitudes; children first observe, remember and try for themselves, then use an AI coach that asks, nudges and encourages verification to create with Chinese, English, code and experiments.
Try first, then use AI; verify, explain and transfer afterwards · Progress is not age-based
Try first ✦
AIcoach4kidToday
What is your evidence?
My first idea
Check and revise
Explain and transfer
Explain it →
Chinese / English
What AI literacy means
More than using AI: growing in a world shaped by it
AI literacy brings together technical knowledge about AI systems, durable skills that transfer across subjects and real problems, and future-ready attitudes. Children learn to work with AI and to critically evaluate its benefits, risks and ethical implications.
Create
Create meaningful work with AI and STEM
Bring subject knowledge, expression, Blockly, experiments and design together to make code, models, prototypes or explanations for a real audience.
Manage
Manage tools, information and judgment
Verify sources and results, protect privacy, understand bias and copyright, and know when to pause or ask a teacher or family member for help.
Shape
Ask how AI should serve people and society
Consider who benefits, who may be left out, and the environmental cost; propose fairer, safer and more responsible improvements.
What an AI system is
A machine-based system that infers outputs from inputs
An AI system is a machine-based system designed to operate with varying levels of autonomy and that may exhibit adaptiveness after deployment. For explicit or implicit objectives, it infers from the input it receives how to generate outputs such as predictions, content, recommendations or decisions that can influence physical or virtual environments.
Inputs and objectives
It receives information and works toward an objective
Inputs may be text, images, sound, sensor data or human actions. Objectives can be explicitly set by people or implicit in a system’s design.
Inference and outputs
It produces predictions, content, recommendations or decisions
An output is a result produced from inputs and objectives. It is not automatically a fact, and it is not the same as human judgment.
Autonomy and adaptiveness
Different systems have different degrees of autonomy
Some systems act only after a human instruction. Others can carry out steps automatically or may exhibit adaptiveness after deployment through data and feedback.
Impact and responsibility
It can affect virtual and physical environments
Recommendations, scoring, navigation, images and generated content can shape people’s choices. We therefore verify results, protect privacy, notice bias and retain human responsibility.
Turning four education challenges into practical learning support
Not another AI tool course: a shared, teachable path for AI literacy
AIcoach4kid connects a common capability definition, everyday contexts, teacher scaffolds and implementation evidence so learners, families and educators share both language and action.
Challenge 01 · No shared language
Use one capability model to define what to learn and teach
Technical knowledge, durable skills and future-ready attitudes become bilingual tasks, observable evidence, assessments and teacher interviews through Engage, Create, Manage and Shape.
Challenge 02 · Inconsistent implementation
Turn strategy into repeatable classroom and school routines
Bilingual content bundles, task templates, teacher learning, process rubrics, family support and staged pilots help classes and schools implement consistently in local contexts.
Challenge 03 · Anxiety and misconceptions
Bring AI back to observable, discussable daily life
Starting with recommendations, translation, images, sorting and school questions, children learn what AI can and cannot do, where it can fail and how not to fear or blindly trust it.
Challenge 04 · Unclear pedagogy
Use AI to enhance learning, never to replace it
Try independently first, then receive a local clue; verify evidence, explain the process and transfer learning. AI does not replace foundational memory, independent thinking or process expression.
AI · Expression · STEM
AI amplifies learning; it does not stand in for an answer
AIcoach4kid connects language expression, science inquiry, mathematical thinking, coding, art and engineering. Children build more than tool fluency: knowledge, questioning, judgment, explanation, transfer and creation.
Knowledge and questions
Build understanding, then make the uncertainty clear
Children first name concepts, examples and guesses in their own words, then turn “I do not get it” into a clarifying, comparison, why, counterexample or transfer question. AI only helps refine the question.
Judgment and STEM inquiry
Move from hypotheses to data, verification and improvement
Children observe, measure, model or prototype, test, compare online and offline results, seek counterexamples and revise conclusions with evidence.
Explanation and transfer
Explain the process, then use it somewhere new
Children explain the question, knowledge, reasoning and checks, then use a concept in a new family, school or community challenge. AI suggestions can be adopted, changed or refused.
Learning guardrails
Protect the cognitive work that belongs to a child
AI can support learning, but it cannot skip learning. The platform separates what AI must not replace from what it can strengthen.
AI does not replace
- Foundational memory
- Independent thinking
- Process expression
AI can strengthen
- Questioning
- Feedback
- Explanation and reflection
- Transfer and creation
A learning loop
Six steps that make learning real
01
Try first
Start from prior knowledge, observation and a first guess; keep the original idea.
02
Make the question clear
Turn uncertainty into a question that can be answered, compared or verified.
03
Explore with feedback
Use Blockly, controlled AI, sources and physical materials to test an idea and get a local clue.
04
Verify evidence
Compare sources, logic, counterexamples, online results and hands-on observations; AI is not the authority.
05
Revise and explain
Explain the question, knowledge, reasoning and checks; record why a suggestion was changed or refused.
06
Transfer and share
Use learning in a new family, school or community situation and explain it to a real audience.
Inside the platform
A learning path grounded in the real world
For families, classrooms, clubs and cross-school projects. Children start from evidence of what they can do, not an age-based level.
Bilingual expression studio
Complete tasks in Chinese, English or side-by-side bilingual mode. Keep original ideas, drafts, AI suggestions, revisions and presentation summaries.
Blockly coding lab
Move from sequence and conditions to loops, variables and models through playful challenges about programs, data and testing.
Online + hands-on experiments
Repeat simulations online, feel the constraints of real materials offline, then compare predictions with observations.
Projects and portfolios
Connect individual learning to audiences and collaboration through roles, project questions, evidence cards, versions and oral defense.
Basic / Intermediate / Advanced
Every level includes an independent assessment and a human interview. Levels describe competencies, not age or grade.
Teachers and families together
Teachers guide safety and professional judgment. Families observe, ask, experiment and reflect without doing the work for children.
Designed for the world
Bilingual is more than a language switch
The interface, lessons, tasks, prompts, experiments, assessments and interviews are available in Chinese and English. Children can think in their strongest language first, then add keywords, summaries or presentations in the other.
Bilingual interfaceBilingual tasksBilingual portfoliosLanguage fairness
中文
My question: How can our school sort waste more accurately?
English
My question: How can our school sort waste more accurately?
AI Literacy progression
Progress has direction, not an age label
Basic, Intermediate and Advanced describe what a learner can do independently now. Each level has an assessment and human interview, with a profile of specific competencies rather than one hidden total score.
Basic
Understand, try, explain and verify with support.
Intermediate
Decompose problems, use evidence, test and iterate independently.
Advanced
Balance trade-offs, respond to audiences and improve systems or prototypes.
Designed for each learning partner
From primary curiosity to secondary projects, with families and educators alongside
The same value direction is adapted to a learner’s experience, language and support needs. It does not set a ceiling by age or use age as a level label.
Primary learners
Begin with observation, rules and expression
Use stories, sorting, movement, blocks and low-risk materials to understand inputs, rules, outputs and simple data; state an idea before receiving a hint.
Secondary learners
Move from evidence and argument to real projects
Compare sources and model results, design measurements and counterexamples, analyze bias and trade-offs, and build cross-subject prototypes or community projects.
Families
Support thinking; do not do the work
Talk about what a child asked, what they did not fully trust and what they changed their mind about; support low-cost home observations and safe boundaries.
Educators
Put AI literacy inside subjects and relationships
Use clear goals, bilingual scaffolds, process evidence and human judgment to connect AI with language, science, mathematics, art, engineering and child protection.
Designed for every learning partner
Children
I have an idea, and I can make it real
Build understanding, ask, verify, explain and create through games, experiments, projects and portfolios.
Teachers
See the original attempt and guide real growth
Orchestrate bilingual tasks, inspect programs, experiments, versions, verification and explanations, and run human assessments and interviews.
Families
Observe, ask and reflect together
Use family activities to ask three questions: What did you ask? What did you not fully trust? What idea did you change? No score comparisons and no doing the work for a child.
Schools
Bring AI literacy into subject projects
Support classrooms, clubs, cross-school collaboration, teacher governance and auditable project delivery.
“
The best AI learning starts with a child’s own question. The platform’s job is to make thinking visible, supported and shareable.
AIcoach4kidLet a child start with their own question
Try a bilingual learning task first, then decide how to grow it at home, in class or across a school project.
Frequently asked questions
Is AIcoach4kid only about using AI?
AI literacy is one part of the platform. Children also practice expression, science inquiry, computational thinking, collaboration, verification and prototype improvement. AI enters only when it has a clear learning purpose.
How do Chinese and English work together?
The interface, content and tasks have bilingual versions. Children can choose Chinese-led, English-led, side-by-side or single-language focus, and keep bilingual titles, summaries and evidence.
Are Basic, Intermediate and Advanced tied to age?
No. Age only informs materials and reading support. Levels are based on specific evidence, and every level requires an assessment and a human interview.
Will AI write answers for children?
The default coach asks questions, offers clues and checks evidence. It does not generate a ready-to-submit essay, complete program or inquiry conclusion. Teachers can inspect the process and guide learning.
What if we do not have materials or AI access?
Lessons include low-cost alternatives, printable task cards and offline drafts. When AI is unavailable, saving, Blockly and local rule-based feedback remain available.