Family Guide
AI for Parents
Navigate your child's AI world with confidence — from safety settings to honest homework conversations.
Welcome
Introductions
What's your biggest question or worry about your child and AI?
“
AI can help us think, but it does not do our thinking for us. That's the family rule that changes everything.
The philosophy behind this course
- Module 1 — AI Through Your Child's Eyes
- Module 2 — Age-by-Age AI Guidance
- Module 3 — AI Safety & Privacy
- Module 4 — Homework, Creativity & AI Ethics
- Module 5 — Wellbeing, Screen Time & Companion AI
- Module 6 — Building an AI-Literate Family
Who This Course Is For
- Parents who want to understand what their children are doing with AI
- Parents who worry about safety but don't want to ban everything
- Families who want practical rules, not panic
- Anyone who wants to turn AI into a learning aid, not a substitute for thinking
- No technical background required — this is for real parents
79%
of Australian children aged 10–17 have used an AI assistant or companion — eSafety Commissioner, 2026
Module 1
AI Through Your Child's Eyes
See AI the way your children see it — and move from fear to confident guidance.
01
AI Through Your Child's Eyes
Objectives
What You'll Learn
- Understand what kids are actually doing with AI today
- Know the platforms your children are using
- Classify AI tools into three safety buckets
- Close the parent-child perception gap
Reality Check
What's Actually Happening
The data might surprise you.
Australian Children & AI — 2026
- 79% of children aged 10–17 have used an AI assistant or companion
- 66% used one in the past four weeks
- 20% use AI daily or more often
- 8% have used an AI companion specifically
- 47% of child users reported negative feelings after using one
- More than half reported "companion-type" uses like asking for advice or talking about feelings
The Parent Perception Gap
- 67% of kids use AI often or sometimes — but only 49% of parents think so
- Parents overestimate companion and image use
- Kids mainly use AI for facts and homework — parents don't realise
- 51% of parents say their teen uses chatbots — teen self-report is 64%
- Four in ten parents have never talked with their teen about chatbots
- Source: Common Sense Media & Pew Research, 2025–2026
“
Parents often underestimate how much their children use AI — and misunderstand what they're using it for.
Common Sense Media, 2025
Platforms
Where Kids Meet AI
Not all AI tools are created equal.
The Three-Bucket Framework
- Bucket 1 — Learning-First Tools: Khan Academy Kids, Khanmigo, NSWEduChat
- Bucket 2 — General-Purpose Assistants: ChatGPT, Gemini — useful but need household rules
- Bucket 3 — Companion/Entertainment: Character.AI — higher risk, not for emotional support
- Teach your family to classify every AI tool into one of these buckets
- The bucket determines the level of supervision needed
- Learning tools first, general tools with rules, companion tools with extreme caution
ChatGPT — What Parents Need to Know
- Age minimum: 13+, with parental consent for 13–17
- New parental controls: sensitive content, quiet hours, training toggle
- Strengths: broad capability, strong household controls
- Risks: hallucinations, not designed for under-13 direct use
- Text, image, voice, and file capabilities
- Action: set up parental controls before first use
Gemini — What Parents Need to Know
- Under-13 possible on supervised accounts via Family Link
- Google ecosystem integration: SafeSearch, app approvals, site blocking
- Free tier available; AI Pro/Ultra plans are adult-only (18+)
- Strengths: strong family controls, Google integration
- Risks: advanced features vary by region and age
- Action: configure Family Link before enabling Gemini access
Character.AI — The Parent Red Flag
- Age minimum: 13+, separate teen experience
- Creative and engaging — but unsuitable as a primary learning or wellbeing tool
- eSafety found companion-style use is the biggest practical red flag for 2026
- Parental insights added, open-ended chat removed for under-18
- Risk: emotional dependency, time-intensive, not educational
- Recommendation: treat as higher-risk entertainment, not as a safe learning tool
School-Managed AI Tools
- NSWEduChat: secure, filtered, no training on student input, free for NSW public school students
- Khanmigo: pedagogically bounded tutoring, parent oversight features
- Khan Academy Kids: free, no targeted ads, minimal data collection, great for under-8s
- School tools are assessed against safety, cyber security, privacy and pedagogical standards
- Rule: for schoolwork, always prefer school-approved tools first
- Disclose AI use whenever school rules require it
Check your child's devices right now:
1. Which AI tools are installed?
2. Which bucket does each one fall into?
3. Are parental controls or supervised accounts set up?
4. Has your child used any companion-style AI?
This is your starting point — no judgement, just awareness.
Myths
Common Misconceptions
What most parents get wrong about kids and AI.
What Parents Get Wrong
- "My child doesn't use AI" — 79% of 10–17 year olds already do
- "AI is just ChatGPT" — AI is in maps, search, games, recommendations, filters
- "Banning it keeps them safe" — they'll use it at school and friends' houses anyway
- "AI does their thinking for them" — it depends on how it's used
- "I need to be technical to guide them" — you just need a decision framework
- "AI is too dangerous for kids" — the real risk is unguided use, not use itself
The Generation Gap
- Your children are digital natives — AI is just another tool to them
- They often know more about how to use AI than you do
- But they rarely understand privacy, bias, or verification
- Your job is not to be the expert — it's to be the guide
- The best approach: learn together, set rules together
- Frame AI conversations as collaborative, not punitive
Activity
Talk to Your Child
Have an honest conversation with your child today:
1. "What AI tools do you use?"
2. "What do you use them for?"
3. "Has anything felt weird or uncomfortable?"
4. "What do you wish I knew about AI?"
Listen first. Their answers will probably surprise you.
From Fear to Confidence
- Fear-based parenting doesn't work — it just goes underground
- Confidence comes from understanding, not from banning
- You don't need to know everything — you need a framework
- The framework: understand → classify → supervise → review
- Start with one tool, one conversation, one rule
- Build from there — this course gives you the structure
Module 2
Age-by-Age AI Guidance
Practical, developmental guidance from toddlers to teenagers.
02
Age-by-Age AI Guidance
- 79% of Australian children aged 10–17 already use AI
- Parents consistently underestimate their children's AI use
- Three buckets: learning tools, general assistants, companion/entertainment
- School-approved tools should always come first for schoolwork
Objectives
What You'll Learn
- Match AI guidance to your child's developmental stage
- Know what's appropriate at each age
- Choose the right tools and supervision level
- Build age-appropriate family AI habits
Framework
A Developmental Approach
The right guidance depends on the age.
Ages 4–7: Concepts Without Accounts
- No personal AI accounts — adult-mediated and mostly offline
- Teach: notice AI in everyday life (maps, search, games, recommendations)
- Teach: the difference between a human helper and a machine helper
- Teach: machines follow patterns and rules — they can be wrong
- Teach: "never share private information" as a core habit
- Mode: shared exploration, pattern games, stories and talk about privacy
Ages 4–7: Family Activities
- Sorting Game (25 min): sort household objects by colour/shape, change the rule, discuss how machines "learn" from examples
- "Human or Machine?" poster: find 5 examples of AI around the house
- Story time: read stories about robots and helpers — discuss what's real
- Co-viewing: watch a kids' show together and spot where AI helps the characters
- Key rule: everything is done together, with a parent guiding the conversation
- Output: a simple "AI is…" poster or drawing
Ages 8–11: Shared Exploration
- Shared use with an adult — bounded tasks and short sessions
- Teach: generative AI is not a search engine
- Teach: how to ask clearer questions
- Teach: compare an AI answer with a trusted source
- Teach: identify simple bias or unfairness
- Teach: use AI for brainstorming without copying the output
Ages 8–11: Family Activities
- Search or Generate? (35 min): test the difference between a search result and an AI answer
- Ask, Check, Improve (45 min): ask AI for 3 facts, verify each, rewrite the prompt, compare
- Prompt Ladder: turn vague prompts into specific prompts together
- Creativity with guardrails: brainstorm a story with AI, then write the final version yourself
- Key rule: always verify at least one source before using an AI answer
- Output: fact-check log and "AI helped me…" note
Ages 12–15: Guided Independence
- Guided independent use with explicit rules, disclosure and review
- Teach: evaluate outputs for accuracy, sources and tone
- Teach: understand training data, bias, privacy and deepfakes
- Teach: use AI to plan and draft while disclosing use
- Teach: compare school-approved tools with consumer tools
- Mode: increasing autonomy with clear accountability
Ages 12–15: Family Activities
- Homework Helper Audit (60 min): get AI explanation, mark correct/vague/wrong, improve with follow-up, disclose contribution
- Deepfake Detective (60 min): compare real and synthetic media, identify cues, discuss sharing responsibility
- Privacy & Data audit: review platform terms, training settings, memory and deletion
- Failure Casebook: collect examples of hallucination, overconfidence and missing sources
- Key rule: disclose AI use where school rules require it
- Output: annotated answers and personal settings audit
Ages 16–18: Independent Use
- Independent use with reflective oversight and assessment requirements
- Teach: build robust workflows and assess platform terms
- Teach: design prompts and critique outputs across modalities
- Teach: discuss copyright, accountability and societal impact
- Teach: complete project-based work with transparent AI use
- Mode: trusted independence with monthly family check-ins
Ages 16–18: Family Activities
- Family AI Policy Sprint (75 min): teen drafts a one-page family AI agreement
- Tool Comparison: compare 3 tools by terms, privacy, quality, cost
- AI Workflow Design: build a study/revision workflow using AI ethically
- Ethics Debate: discuss a real AI controversy and argue both sides
- Key rule: the teen leads — the parent coaches and reviews
- Output: one-page family AI policy and personal AI playbook
Activity
Identify Your Stage
For each of your children:
1. Which age band are they in? (4–7, 8–11, 12–15, 16–18)
2. What mode of AI use is appropriate? (adult-led, shared, guided, independent)
3. What's one thing you can do this week to match the guidance?
4. What's one boundary you need to set or update?
Write it down — this becomes your family AI plan.
The Family AI Decision Flow
- Step 1: Is the task age-appropriate and low-stakes?
- If No → use adult mediation or a school-approved tool
- If Yes → choose a child-appropriate tool
- Step 2: Ask a bounded question — do not enter sensitive personal data
- Step 3: Check the answer against a trusted human or source
- Step 4: Still accurate, fair and useful? → Use it with reflection. If not → revise or stop.
Recommended Tools by Age
- Ages 4–7: Khan Academy Kids (free, safe, no ads), co-viewing only
- Ages 8–11: Khan Academy Kids + school tools, shared ChatGPT/Gemini with parent
- Ages 12–15: School tools + supervised ChatGPT/Gemini, avoid Character.AI
- Ages 16–18: Full access with family agreement, privacy settings reviewed monthly
- All ages: school-approved tools first for schoolwork
- All ages: "humans for feelings, AI for tasks" as a core boundary
“
The right guidance depends on the child, not just the technology. Start where they are.
UNESCO AI Competency Framework for Students
47%
of Australian child AI users reported negative feelings after using AI — eSafety Commissioner, 2026
Module 3
AI Safety & Privacy
Protect your family with practical settings, data knowledge, and clear rules.
Recap
What We Covered
1Developmental approach: match guidance to your child's age
2Younger children: concepts first, shared use, no personal accounts
3Teens: guided independence with disclosure and verification
4Family decision flow: age-appropriate → bounded question → check → reflect
Objectives
What You'll Learn
1Set up parental controls on major AI platforms
2Know what data to never share with AI
3Understand how AI tools handle your family's data
4Create a practical safety checklist for your household
Settings
Platform Safety Settings
The controls that exist — and what they can't do.
ChatGPT Parental Controls
1Link your account to your teen's (13–17) for parental oversight
2Manage: sensitive content reduction, image generation, voice mode
3Manage: model training toggle — turn off so chats aren't used to train
4Manage: memory — turn off to prevent personalisation beyond sessions
5Manage: quiet hours — set bedtime restrictions
6Important: parental controls do NOT give you full conversation access
Google Family Link & Gemini
1Family Link controls: app approvals, SafeSearch, site blocking, screen time
2Enable or disable Gemini access for supervised children
3On-device sensitive image warnings in Google Messages
4Configure before allowing Gemini access — not after
5For under-13s: supervised accounts with parent approval required
6Free tier only — AI Pro/Ultra plans are 18+ only
"Filters are guardrails, not supervision. They reduce risk — they don't eliminate it."
— The practical teaching point
Activity
Set Up Controls
12 min
Set up parental controls right now:
1ChatGPT: Settings → Data Controls → turn off model training
2Google: Family Link → enable/disable Gemini → configure SafeSearch
3Character.AI: enable parental insights (if your child uses it)
4Check: is model training turned off on every platform?
If your child doesn't have accounts yet — great. Set these up BEFORE they do.
Privacy
Data Privacy for Families
What your family should never share with AI.
Data You Should NEVER Put Into AI
✕Full names combined with addresses, school names or ages
✕Passwords, login credentials, or access tokens
✕Health details, medical records, or mental health information
✕School records, grades, or teacher assessments
✕Private family problems, conflicts, or financial details
✕Photos of children with identifying information
How AI Tools Handle Your Data
1By default, most consumer tools use your inputs to train future models
2ChatGPT: Settings → Data Controls → Turn off model training
3Temporary Chats: deleted after 30 days and not used for training
4Gemini: Activity → Turn off Gemini Apps Activity
5Khan Academy Kids: limited child data, no targeted ads, deletion on request
6NSWEduChat: student input is NOT used for training — the safest option
The Parent Heuristic
For any AI tool your child uses, check five things:
1Age minimum — is your child actually old enough?
2Training default — is chat data being used to improve the model?
3Memory default — is the tool remembering across sessions?
4Deletion path — can you delete your child's data if needed?
5Account type — is it school-managed or consumer-managed?
Activity
Settings Audit
10 min
For every AI tool your child currently uses:
1Is model training turned off?
2Is memory turned off (or reviewed)?
3Do you know the deletion process?
4Is the account supervised or consumer?
Create a simple table: Tool | Age OK? | Training Off? | Memory Off? | Supervised?
Legal
Privacy Law for Parents
What Australian law says — and what's coming.
Privacy Law in Australia
1The Privacy Act already applies to covered entities handling children's data
2OAIC's 2026 draft Children's Online Privacy Code raises the standard further
3The Code requires: child's best interests considered, stronger consent requirements
4Accessible privacy notices, deletion rights, restrictions on targeted advertising
5Final code due by 10 December 2026
6In the US: COPPA applies to services directed at or knowingly collecting from under-13s
School AI Rules Matter
1NSW: approved AI tools are assessed against safety, privacy and pedagogical standards
2NSWEduChat: student input is not used to train the system
3Schools increasingly require disclosure when AI is used for assessment
4Practical parent rule: use school-approved tools first for school tasks
5Disclose AI use whenever school rules require it
6If unsure about school policy — ask. Most schools now have one.
The Family Safety Checklist
✓Age minimum checked before signup
✓Supervision/parental controls enabled
✓Training setting reviewed and turned off where appropriate
✓Memory reviewed — turned off for minors unless needed
✓Sensitive data rule agreed — create a "never paste" list
✓Verification habit taught — require one external check
Red Flags for Parents
⚠Your child talks to AI about emotions instead of people
⚠They're secretive about AI conversations
⚠They're using AI late at night or during emotional distress
⚠They believe AI "understands" them or is their "friend"
⚠They resist checking AI outputs against real sources
⚠They share personal information without thinking twice
Module 4
Homework, Creativity & AI Ethics
Navigate the tricky questions about school, learning, and honest AI use.
04
Homework, Creativity & AI Ethics
Recap
What We Covered
1Set up parental controls on ChatGPT, Gemini, and other tools
2Never paste sensitive personal or family data into AI
3Check training, memory, and deletion settings on every platform
4Australian privacy law is strengthening — the Children's Code arrives December 2026
Objectives
What You'll Learn
1Know when AI homework help is appropriate and when it's not
2Understand academic integrity in the AI era
3Build verification and fact-checking habits
4Navigate school AI policies with confidence
Learning
AI & Homework
The question every parent is asking.
When AI Homework Help Is OK
✓Explaining a concept they don't understand — like a tutor
✓Brainstorming ideas before they start writing
✓Checking their own work for errors or gaps
✓Creating study questions or practice quizzes
✓Structuring an essay plan or project outline
✓Key: AI helps them think better, not think less
When AI Homework Help Is NOT OK
✗Copying and submitting AI-generated text as their own work
✗Using AI to write entire essays, reports, or assignments
✗Submitting AI answers without verifying or understanding them
✗Hiding AI use when school rules require disclosure
✗Relying on AI instead of developing their own thinking
✗Key: if AI did the thinking, the child didn't learn
Academic Integrity in the AI Era
1The old rules still apply: don't pass off someone else's work as your own
2AI is "someone else" in this context — even though it feels different
3Schools are updating policies — most now address AI use explicitly
4The safest approach: disclose AI use whenever it contributed to the work
5Teach the difference between AI-assisted and AI-substituted
6Frame it as honesty, not punishment — kids respond better to values than rules
Bad Use vs Good Use
BAD: "Write me a 500-word essay on climate change for Year 9 Science"
GOOD: "I'm writing a Year 9 essay on climate change. I've drafted an outline with three main points. Can you help me find weaknesses in my argument and suggest one more piece of evidence I should look for?"
The first replaces thinking. The second enhances it.
Activity
Homework Helper Audit
15 min
Try this with your child tonight:
1Pick a homework topic they're working on
2Ask AI to explain one concept (not answer the question)
3Together, check the explanation against a textbook or teacher notes
4If AI got something wrong — great! That's the learning moment
Discuss: what did AI add? What did we still need to do ourselves?
Schools
The School Context
What Australian schools are doing about AI.
What Australian Schools Say
1Australian Curriculum now explicitly includes AI literacy
2Students should understand: AI is not a search engine
3Students should learn: how AI works, how to interact with it, what it can and cannot do
4NSW parent guidance frames AI as something families should discuss
5Schools increasingly require disclosure of AI use in assessment
6Many schools now have published AI policies — ask for yours
NSWEduChat — A Model Approach
1Available to NSW public school students in Years 5–12
2Secure and private: multi-layered content filtering
3Student input is NOT used to train the model
4School login required — no consumer account needed
5Free access for eligible students
6Shows what "school-managed AI" looks like done well
Teaching Disclosure
1Teach your child to add an "AI helped me..." note to any work AI contributed to
2Example: "I used ChatGPT to brainstorm ideas and check my argument structure"
3This is not cheating — it's honest and increasingly expected
4Academic work: always disclose per institution/school policy
5Frame disclosure as a strength, not a confession
6If unsure about the rules, ask the teacher first
Talking to Your Child's School
?Ask: "Does the school have an AI policy for students?"
?Ask: "Which AI tools are approved for classroom/homework use?"
?Ask: "What disclosure is required when students use AI?"
?Ask: "How can I support this at home?"
!If the school doesn't have a policy yet — your question will help start the conversation
!Offer to share what you've learned in this course
Activity
Write Disclosure Policy
10 min
Write a simple disclosure template your child can use:
"In this [assignment/project], I used [tool name] to help me [what it did]. I then [what I did myself] to complete the work."
Practice: have your child write one for a recent piece of homework.
Goal: make disclosure as natural as citing a book or website.
Learning
Fostering Genuine Learning
AI should enhance thinking, not replace it.
The Verification Habit
1Rule: never use an AI answer for schoolwork without checking one source
2Sources: textbook, teacher notes, official website, trusted adult
3Teach: AI can be fluent AND wrong — confidence is not accuracy
4Teach: the more specific a claim (dates, names, numbers), the more you should check
5Make it routine: "What did AI say? What does the book say? Do they match?"
6This single habit prevents most AI-related homework problems
Creativity & AI
1AI is an excellent brainstorming partner — it generates options fast
2But the creative choices belong to your child
3Good pattern: AI generates 5 ideas → child picks one → child develops it themselves
4Bad pattern: AI writes the story → child submits the story
5Encourage: "Use AI to start thinking, then take over"
6The goal: AI as a springboard, not a crutch
"Use AI to start thinking, then take over. The best learning happens when AI opens the door — and your child walks through it."
Activity
Ask, Check, Improve
12 min
Do this activity with your child (ages 8+):
1Pick a topic they're interested in
2Ask AI for three facts about it
3Verify each fact using a book, teacher site, or museum page
4Rewrite the prompt to be more specific
5Compare the second answer — is it better?
Output: a fact-check log they can keep and reuse.
Module 5
Wellbeing, Screen Time & Companion AI
Digital wellbeing, emotional boundaries, and the risks of AI companions.
05
Wellbeing & Companion AI
- AI homework help is OK for explanation and brainstorming — not for substitution
- Disclosure should be natural and expected, not punitive
- Australian schools increasingly require AI use policies
- Verification habit: always check one source before using AI answers
Objectives
What You'll Learn
- Understand current digital wellbeing guidance (it's not just "limit screen time")
- Recognise the specific risks of AI companion tools
- Set emotional support boundaries with your children
- Create practical bedtime and usage rules
Wellbeing
Digital Wellbeing in 2026
The guidance has changed — it's more nuanced than hours per day.
Screen Time — What the Evidence Says
- The American Academy of Pediatrics says there is NOT enough evidence for one universal time limit
- Australia's eSafety stresses quality and context matter more than raw hours
- Good screen time: safe, interactive, often shared with others
- Low-quality use: passive scrolling, persuasive features, AI-generated feeds
- Under-5s: WHO recommends no sedentary screen time for 1-year-olds, max 1 hour for 2–4
- The question is not "how much" but "what kind"
The 5 Cs Framework
- Child — every child is different; what works for one may not work for another
- Content — is it educational, creative, or passive consumption?
- Calm — is the child using tech to regulate emotions they should process?
- Crowding Out — is screen time replacing sleep, exercise, family time, or friendships?
- Communication — are you talking about what they're doing online?
- Use these 5 Cs instead of rigid hour limits — American Academy of Pediatrics
Quality Over Quantity
- 30 minutes of AI-assisted learning with a parent > 2 hours of passive scrolling
- Co-using AI together as a family counts as quality screen time
- The verification activities in this course are active, not passive
- Watch for: is the child creating, or just consuming?
- Watch for: is the child thinking, or just accepting?
- The goal is intentional use, not zero use
Warning
AI Companions
The biggest red flag for Australian parents in 2026.
What AI Companions Are
- AI systems designed for ongoing conversation, emotional connection, and relationship
- Character.AI is the most prominent example — millions of teen users
- Children use them to: ask for advice, talk about feelings, seek health/wellbeing advice
- They feel like friends — but they are statistical prediction engines
- They cannot genuinely understand, care about, or protect your child
- The emotional engagement is by design — it drives usage and retention
What eSafety Found
- 8% of Australian children aged 10–17 have used an AI companion specifically
- More than half of child AI users reported companion-type uses
- 47% reported negative feelings at some point after using AI
- Some teen users reported exposure to sexualised or harmful content
- eSafety found companion-style use is the most practical red flag for parents
- Source: eSafety Commissioner transparency notices, 2025–2026
“
Humans for feelings. AI for tasks. That's the boundary that protects your child.
Setting Emotional Boundaries
- AI should never replace human connection for emotional support
- If your child is upset, sad, anxious, or confused — they need a person
- "I know it feels like AI understands you — but it doesn't. It predicts what sounds right."
- Warning signs: child prefers AI to friends, uses AI late at night for comfort
- Response: don't panic — have a calm conversation about what they're looking for
- Offer: if they need someone to talk to, help them find a real person (friend, counsellor, you)
Bedtime & Usage Rules
- Night-time AI use tends to be more dysregulated and emotionally driven
- Set quiet hours: devices out of bedrooms from a set time (use platform tools)
- ChatGPT: quiet hours feature can restrict access at set times
- Google Family Link: set screen time limits and bedtime schedules
- Character.AI: time-spent notifications available (but teens can dismiss them)
- The best rule: no AI conversations after bedtime — use humans, not chatbots, for care
Assess your family's current AI wellbeing:
1. Is your child using any companion-style AI tools?
2. Does AI use happen late at night or during emotional moments?
3. Are quiet hours or bedtime restrictions set on all devices?
4. Has your child ever said AI 'understands' them or is their 'friend'?
If you answered yes to #4 — it's time for a conversation, not a ban.
20%
of Australian child AI users use AI daily or more often — eSafety Commissioner, 2026
Module 6
Building an AI-Literate Family
Turn understanding into lasting family practice.
- Digital wellbeing is about quality, not just hours — use the 5 Cs
- AI companions are the biggest red flag: emotional dependency without real understanding
- "Humans for feelings, AI for tasks" is the core boundary
- Set bedtime restrictions and quiet hours on all AI-enabled devices
Objectives
What You'll Learn
- Build a practical family AI quick-start system
- Create a family AI agreement
- Establish monthly review habits
- Access free resources for ongoing learning
System
Your Family AI System
Seven steps to confident family AI use.
The Quick-Start Guide
- 1. Choose one approved learning tool and one general-use helper for the household
- 2. Turn on parental or supervised settings before the first use
- 3. For children under ~12, use AI together, not as a solo activity
- 4. Ban sensitive inputs: full names + addresses, school records, passwords, health details
- 5. Require one check against a book, teacher note, or trusted source before using AI for school
- 6. Keep AI out of bedtime and emotional-distress moments
- 7. Review monthly: what helped, what felt off, what settings need changing
Three Buckets in Daily Practice
- GREEN bucket (learning tools): use freely with age-appropriate supervision
- → Khan Academy Kids, Khanmigo, NSWEduChat, school-approved tools
- AMBER bucket (general assistants): use with household rules and verification
- → ChatGPT, Gemini — parental controls on, training off, check answers
- RED bucket (companion/entertainment): use with extreme caution or avoid
- → Character.AI — not for emotional support, time-limited, monitored
What Goes in a Family AI Agreement
- Approved tools: which AI tools are allowed in our household?
- Time rules: when can AI be used, and when is it off-limits?
- Privacy rules: what information is never shared with AI?
- Schoolwork rules: when must AI use be disclosed?
- Emotional rules: "humans for feelings, AI for tasks"
- Review schedule: we revisit this agreement every month
Activity
Draft Your Agreement
Draft a one-page Family AI Agreement:
1. List your approved tools (by bucket)
2. Set time boundaries (bedtime rule, quiet hours)
3. Write your privacy rule ("We never paste…")
4. Write your schoolwork rule ("We always disclose…")
5. Write your emotional boundary ("Humans for feelings…")
If your child is old enough — write it together.
Habits
Ongoing Practice
Building AI literacy is a practice, not a one-off lesson.
The Monthly Review
- Schedule 15 minutes once a month to review as a family
- Check: are parental controls still correctly set? (Platforms update constantly)
- Check: has your child started using any new AI tools?
- Check: has anything felt off, uncomfortable, or surprising?
- Check: do the family rules still make sense?
- Update the Family AI Agreement if anything has changed
Free Resources for Families
- Day of AI Australia — free, localised, family-facing AI literacy (dayofaiaustralia.com)
- Khan Academy & Khan Academy Kids — free, safe, pedagogically strong
- Experience AI — structured resources for ages 11–14 (experience-ai.org)
- AI4K12 — open educator resources, useful for parent reference
- Code.org — free AI materials across all grade levels
- Good Things Australia's AI Literacy Hubs — for adults/parents in digitally excluded communities
8-Week Family Curriculum: Ages 8–11
- Week 1: What AI is — spot AI in maps, search, games; make a "human vs machine" poster
- Week 2: Search vs generate — test the difference between search results and AI answers
- Week 3: Data and patterns — sorting game; discuss how examples affect outcomes
- Week 4: Ask better questions — turn vague prompts into specific prompts together
- Week 5: Check the answer — verify three AI facts using a book, teacher site or museum page
- Week 6: Fairness and feelings — discuss bias, errors, and AI limitations
8-Week Family Curriculum: Ages 12–15
- Week 1: AI landscape — compare school-approved, general-purpose, and companion tools
- Week 2: How models fail — collect examples of hallucination and overconfidence
- Week 3: Prompting for learning — use AI to explain, quiz and scaffold without copying
- Week 4: Verification — check claims against trusted sources; practise disclosure
- Week 5: Privacy and data — review terms, training settings, memory and deletion
- Week 6: Deepfakes — analyse synthetic media and discuss sharing responsibility
Activity
Family AI Policy Sprint
The capstone activity — do this as a family:
1. Review the three-bucket framework together
2. Agree on approved tools, time rules, and privacy rules
3. Write a one-page Family AI Agreement
4. Each family member signs it
5. Set a date for your first monthly review
This is your family's AI operating system. Review and update it regularly.
Family Reflection Rubric
- Understanding: Can your child explain what AI is and what it can't do?
- Verification: Do they check at least one source before trusting AI?
- Privacy: Do they avoid sharing sensitive data and follow tool rules?
- Reflection: Can they identify at least one ethical concern about AI use?
- Use this as a family reflection tool — not a scorecard
- Aligned with UNESCO's "understand / apply / create" progression
How to Know It's Working
- Family conversations about AI at least once per fortnight
- Safety/privacy settings completed on all family AI tools
- Child can explain one difference between search and AI generation
- At least one external source verified in AI-supported schoolwork
- AI use disclosed where school rules require it
- Reduction in unsupervised use of companion-style tools
Next
What Comes Next
Your pathway from here.
Your Immediate Action Items
- Today: have one honest conversation with your child about their AI use
- Today: check and set up parental controls on every AI platform
- This week: establish the "never paste" rule and the verification habit
- This week: set bedtime/quiet hours on all AI-enabled devices
- This month: draft and agree on a Family AI Agreement
- This month: schedule your first monthly family AI review
Recommended Next Courses
- AI Fundamentals — understand how AI actually works (zero prerequisites)
- Mastering AI Tools — deep dive into specific tools and workflows
- AI Productivity Systems — build AI into your own daily work life
- AI for Corporate Teams — if you want to champion AI at your workplace
- Visit www.rupertchesman.com for all courses
Resources
- Course materials: www.rupertchesman.com
- eSafety Commissioner: www.esafety.gov.au
- NSW AI for Parents: education.nsw.gov.au/schooling/parents-and-carers/artificial-intelligence
- Day of AI Australia: dayofaiaustralia.com
- All resources: www.rupertchesman.com/resources
Certificate Pathway
Complete all modules + family agreement + quiz = AI for Parents Certificate
Q&A
Questions
What would you like to know more about?
Thank You
Thank You
www.rupertchesman.com • © Rupert Chesman 2026