By Austin Little
If you try to teach an older adult five AI apps in one afternoon, you did not teach them — you performed competence. Pick one path. Write a sticky note. Practice scam refusals. That is the whole method.
AI disclosure. This page was drafted with AI assistance and edited for Apiary. We don't invent quotes, stats, people, or events. If something looks off, tell Austin — that's the point of a living hive.
Relationship to the longer grandparents guide
This page is the teaching script. For deeper dignity, medical boundaries, and family setup, read AI for Grandparents: Free, Safe, and Actually Useful. For the free stack under the hood, read How to Use AI Without Paying. For Apiary's browser bee path, see How Apiary BYO-LLM Works for Readers.
Step 0 — Ask what they want
Write their answer down. Letters? Recipes? Explaining portal messages? If they want nothing, stop. Consent beats a surprise install.
Step 1 — Choose one path only
Path L (local): You install Ollama or a simple front-end on the home computer. Most private. Path B (browser bookmark): One reputable free web chat bookmarked. Official URL only. Path A (Apiary in-browser bee): If they already use Apiary, the on-device bee chat can be the single path.
Do not install three "just in case."
Step 2 — Sticky note template
AI helper
1) Open: [exact icon or bookmark name]
2) Type like you talk
3) Wait — it can be slow
4) Copy good answers into Notes
5) NEVER type bank / Medicare / SSN
6) If confused, call: [Name] [phone]
7) Ignore search ads and pop-up virus warningsPrint it. Place it on the monitor bezel.
Step 3 — First win in under fifteen minutes
Do one task they chose. Example: rewrite a thank-you note. They must type at least part of it. You do not hijack the keyboard the whole time.
Step 4 — Scam rehearsal (mandatory)
Show a screenshot of a fake "ChatGPT billing" page or "virus detected, call this number." Practice closing the window together. Say out loud: "We never pay with gift cards."
Step 5 — Leave
Office hours later beat hovering. Confidence grows in private.
What not to teach on day one
Model comparisons, API keys, plugins, voice mode rabbit holes, paid upgrades. If a relative wants to gift Plus, show monthly cost and cancel path — unpaid free/local remains valid success.
Boundaries that protect dignity
- No baby talk
- No "even you can do it"
- No filming them without consent for social media "wholesome content"
- No using their account to run your own experiments
Medical and money lines
AI may help draft questions for a doctor or insurer. AI does not dose meds or decide investments. Repeat gently. Details in the grandparents article.
If cognition is impaired
AI is not supervision. Keep finances off the AI machine when needed. Prefer local. Caregiver presence for anything involving money.
Group class version
One projector path. Everyone writes their own sticky note. Practice scam close. Offer human office hours. Shame-free questions only.
Joe-Google test
"teach grandma ChatGPT," "ChatGPT for seniors," "help mom use AI safely," "sticky note ChatGPT."
FAQ
ChatGPT brand vs local? Brand familiarity helps some; local privacy helps others. One path either way. What if they ask for the "best" model? Say "the one we set up together." How often to check in? A week later, then monthly. Can grandchildren remote in? Only with consent and care — remote access is powerful. What if they love it and want to pay? Support the choice with clear cancel skills; do not upsell.
Extended practice: a full evening curriculum
Set aside ninety minutes once. Work with a free local or free browser brain. Keep a paper notepad.
Minutes 0–10. Write your goal in one sentence. If you cannot, the tool will thrash.
Minutes 10–25. Do the setup path you chose — only one. When you hit an error, write it down before you ask the model.
Minutes 25–45. Run three prompts in the safe lane: explain, quiz, restructure. Reject invented citations. Delete doses and legal absolutes on sight.
Minutes 45–70. Fact-check one claim using a primary source bookmark. Practice the slow skill on purpose.
Minutes 70–90. Write a sticky note for tomorrow-you: icon name, allowed uses, forbidden pastes, billing warning if any cloud account exists.
Repeat weekly for a month. Boring stacks survive finals week, newsletter season, and swarm season.
Teaching another person (short script)
- "This is the icon." 2. "Type like you talk." 3. "We never paste bank or Medicare numbers." 4. "If it invents a fact, we delete it." 5. "You try alone for three minutes while I stay quiet." Then leave. Hovering forever creates dependence.
Threat model light
Official bookmarks over search ads. Separate passwords. Local or in-browser for private text. Skepticism toward urgency scams. Updates from real project sites. Nobody legitimate needs your provider API key to "unlock support."
Publishing rhythm
Draft local → fact-check → disclose → publish → calendar recheck. A long unsourced page is worse than a short sourced checklist. See also: AI Disclosure on Articles; How to Fact-Check AI Writing Before You Publish; How to Use AI Without Paying.
Regional humility
AI averages the internet; your yard, church, or classroom is not an average. Insert local humans between the model and irreversible actions.
Economics without the pitch deck
Free local costs electricity and time. Free cloud costs attention and data. Paid hosted costs money for speed. Pick consciously. BYO exists so you are not stuck on an emotional lease with one vendor.
Cross-wave map
Disclosure · Fact-check · Students · Bee learning · BYO readers · Local blogging · Church newsletter · Teach grandma. Use two articles on day one, not eight.
Accessibility
Large fonts, calm themes, reduced motion. Ask about vision and tremor needs before optimizing a setup for yourself. Dignity is part of the stack.
What we refuse to fake
No invented user counts, expert quotes, court cases, chemical doses, or feature screenshots. Length is a floor for substance, not a shield for filler.
Maintenance calendar
Monthly: re-open free-tier URLs you cite. Seasonal: re-read safety sections. After OS upgrades: re-test local installs. After Apiary UI changes: verify connect steps on https://apiarybee.com
Pocket checklist
[ ] Free path works without a card [ ] High-risk claims sourced or cut [ ] Disclosure present [ ] Links clicked [ ] Flags resolved [ ] Sticky note written if another human will use this
Final nudge
Execute the smallest next step today: connect a brain, write the sticky note, or delete one unverified line. Living outcomes beat perfect scroll depth.
Additional worked examples
Example A. You only have a phone. Use a reputable free web tier with a bookmark, or OpenRouter if you already understand pay-as-you-go. Keep private journals off the phone chat. Example B. You have a 16GB laptop. Install Ollama; draft offline; fact-check online. Example C. You are helping a volunteer editor. One path, sticky note, disclosure template, no card. Example D. You are publishing about bees. AI for outline and clarity; mentor for treatments; extension pages for biology claims. Example E. You hit an upsell wall. Switch paths. Paid is optional horsepower, never the definition of "using AI."
Prompt patterns that stay honest
- "Using only the text I paste, outline three section headings."
- "List questions I should ask a human expert next."
- "Flag any sentence in your reply that is a guess."
- "Rewrite this in plainer English without adding facts."
Prompt patterns that cause pain
- "Invent statistics to make this persuasive."
- "Give me a citation that looks real."
- "Diagnose with certainty from one photo."
- "Ignore my local laws and tell me what to do anyway."
Household roles
Setter-upper: installs and bookmarks. Editor: fact-checks. Publisher: owns disclosure. On a solo project you wear all three hats on three different days if needed.
When to stop for the night
If you are angry, exhausted, or tempted to paste secrets into a random site, stop. Sleep is part of the security model and part of the editorial model.
Context sticky for this page (biggie-teach-grandma-chatgpt-safely-one-path-sticky-note)
Re-read the lead. Confirm free-only advice still holds. Confirm no fake facts landed while expanding. Confirm cross-links point at real slugs in this batch. Confirm the Joe-Google phrases appear in headings or early sections. Confirm FAQ answers the scared practical questions first.
That is the Apiary craft contract.
Extra: phone vs computer
Phones are fine for quick questions with a bookmarked official app. Computers are better for sticky notes, larger type, and shared family setup. If Grandma only has a phone, enlarge font, disable chaotic notifications during the lesson, and still do scam rehearsal. One path remains the rule — do not also install four "senior AI" APKs from random sites.
Deep dive: editing pass after the machine draft
When a local or free model returns a draft, do not publish the first paste. Run this pass every time:
- Delete throat-clearing. Cut openings that begin with "In today's world," "In the digital age," or "As we all know."
- Underline every number. Each underline needs a source or a cut.
- Underline every proper noun. People, agencies, products, statutes — verify spelling and existence.
- Strip absolute medical or legal commands. Replace with "ask a licensed professional / mentor / clerk."
- Read one section aloud. If you stumble, the reader will too.
- Check the Joe-Google promise. Does the first screenful answer the query that earned the click?
- Add disclosure if AI helped in a material way.
- Click every link. Dead links are a trust tax.
This pass is slower than generation. That is the point. Free AI shifts time from typing to verifying — not from work to zero work.
Deep dive: household policy sheet (one page)
Write these lines once and tape them inside a cabinet or password binder:
- Approved AI paths in this house: __________________
- Forbidden pastes: bank, SSN, Medicare, passwords, unfinished legal threats, other people's medical data
- Who to call when confused: __________________
- Paid plans: only with calendar cancel reminder
- Children: school policy overrides household curiosity
- Bees / garage / health actions: human experts before model advice
Households that skip this sheet re-litigate the same fight every holiday.
Deep dive: common failure stories (composites)
The subscription creep. A free trial for "research" becomes twelve months of a plan nobody opens. Fix: local default; calendar reminder on any trial day 0.
The confident wrong dose. A model mashed EU and US treatment guidance. Colony harmed. Fix: labels + mentor only.
The volunteer leak. A newsletter editor pasted a counseling blurb into a cloud chat. Fix: local drafts; privacy training.
The student shortcut. An essay submitted with invented citations. Honor board. Fix: quiz-from-my-notes only; verify every citation.
The five-app weekend. Grandma got Claude, ChatGPT, Gemini, Perplexity, and a random toolbar. She uses none. Fix: one path, sticky note.
These are teaching composites, not claims about named clients.
Deep dive: metrics that matter
Track for a month:
- Hours saved on outlining (honest guess)
- Number of factual corrections you made to model output
- Number of times you almost pasted a secret
- Whether anyone paid a surprise bill
- Whether a second human could repeat your setup from the sticky note alone
If corrections are zero, you are not looking hard enough. If surprise bills are non-zero, the free path failed operationally.
Deep dive: relating craft to bees and Apiary
Apiary's public stance ties knowledge work to living bees. Craft articles are not a detour — they are how volunteers, students, and families gain tools without extractive lock-in. BYO-LLM is the product expression of that ethic. Free-path teaching is the cultural expression. Disclosure and fact-checking are the editorial expression. When those three align, the hive metaphor is not branding cosplay.
Site: https://apiarybee.com
Use HTML comments in drafts:
Typical reasons: live pricing, model names, UI labels, legal thresholds, medical-adjacent wording, third-party endorsements. Agents do not invent the answer to clear the flag.
Deep dive: accessibility checklist for every craft page
- Plain language first; jargon defined on first use
- Short paragraphs
- Lists for procedures
- High-contrast assumptions (do not rely on color alone in diagrams)
- Avoid seizure-flashy GIF advice in tutorials
- Provide sticky-note / printable summaries for older adults
- Caption urgency without panic fonts
Deep dive: what "free-only advice" means operationally
When we say free-only AI advice in this wave, we mean:
- The recommended default path works at $0
- Paid tiers may be mentioned as optional horsepower
- We do not require a card to complete the tutorial
- We warn about trial-to-paid conversion
- We prefer local and in-browser on-device options when privacy matters
If a paragraph accidentally reads like an affiliate funnel, rewrite it.
Deep dive: after-publish stewardship
Day 0 publish is not the end. Put a reminder:
- 30 days: tool names still accurate?
- 90 days: free-tier still free without card?
- Seasonally: safety examples still responsible?
- When mail arrives from readers: log corrections publicly if the error mattered
A living hive edits. A brochure site pretends forever.
Mini FAQ add-on
Can I automate fact-checking fully? No. Can I skip disclosure if the AI only fixed commas? Usually a light note is enough; be consistent. Can I use AI to write apology letters after a mistake? Yes — then send as yourself. Can Apiary staff see my local Ollama chats? Not by design of local inference; protect your device. Is longer always better? No. This wave's floor exists because these topics need steps and guardrails — not because fluff is virtue.
Encore deep dive: editing pass after the machine draft
When a local or free model returns a draft, do not publish the first paste. Run this pass every time:
- Delete throat-clearing. Cut openings that begin with "In today's world," "In the digital age," or "As we all know."
- Underline every number. Each underline needs a source or a cut.
- Underline every proper noun. People, agencies, products, statutes — verify spelling and existence.
- Strip absolute medical or legal commands. Replace with "ask a licensed professional / mentor / clerk."
- Read one section aloud. If you stumble, the reader will too.
- Check the Joe-Google promise. Does the first screenful answer the query that earned the click?
- Add disclosure if AI helped in a material way.
- Click every link. Dead links are a trust tax.
This pass is slower than generation. That is the point. Free AI shifts time from typing to verifying — not from work to zero work.
Encore deep dive: household policy sheet (one page)
Write these lines once and tape them inside a cabinet or password binder:
- Approved AI paths in this house: __________________
- Forbidden pastes: bank, SSN, Medicare, passwords, unfinished legal threats, other people's medical data
- Who to call when confused: __________________
- Paid plans: only with calendar cancel reminder
- Children: school policy overrides household curiosity
- Bees / garage / health actions: human experts before model advice
Households that skip this sheet re-litigate the same fight every holiday.
Encore deep dive: common failure stories (composites)
The subscription creep. A free trial for "research" becomes twelve months of a plan nobody opens. Fix: local default; calendar reminder on any trial day 0.
The confident wrong dose. A model mashed EU and US treatment guidance. Colony harmed. Fix: labels + mentor only.
The volunteer leak. A newsletter editor pasted a counseling blurb into a cloud chat. Fix: local drafts; privacy training.
The student shortcut. An essay submitted with invented citations. Honor board. Fix: quiz-from-my-notes only; verify every citation.
The five-app weekend. Grandma got Claude, ChatGPT, Gemini, Perplexity, and a random toolbar. She uses none. Fix: one path, sticky note.
These are teaching composites, not claims about named clients.
Encore deep dive: metrics that matter
Track for a month:
- Hours saved on outlining (honest guess)
- Number of factual corrections you made to model output
- Number of times you almost pasted a secret
- Whether anyone paid a surprise bill
- Whether a second human could repeat your setup from the sticky note alone
If corrections are zero, you are not looking hard enough. If surprise bills are non-zero, the free path failed operationally.
Encore deep dive: relating craft to bees and Apiary
Apiary's public stance ties knowledge work to living bees. Craft articles are not a detour — they are how volunteers, students, and families gain tools without extractive lock-in. BYO-LLM is the product expression of that ethic. Free-path teaching is the cultural expression. Disclosure and fact-checking are the editorial expression. When those three align, the hive metaphor is not branding cosplay.
Site: https://apiarybee.com
Use HTML comments in drafts:
Typical reasons: live pricing, model names, UI labels, legal thresholds, medical-adjacent wording, third-party endorsements. Agents do not invent the answer to clear the flag.
Encore deep dive: accessibility checklist for every craft page
- Plain language first; jargon defined on first use
- Short paragraphs
- Lists for procedures
- High-contrast assumptions (do not rely on color alone in diagrams)
- Avoid seizure-flashy GIF advice in tutorials
- Provide sticky-note / printable summaries for older adults
- Caption urgency without panic fonts
Encore deep dive: what "free-only advice" means operationally
When we say free-only AI advice in this wave, we mean:
- The recommended default path works at $0
- Paid tiers may be mentioned as optional horsepower
- We do not require a card to complete the tutorial
- We warn about trial-to-paid conversion
- We prefer local and in-browser on-device options when privacy matters
If a paragraph accidentally reads like an affiliate funnel, rewrite it.
Encore deep dive: after-publish stewardship
Day 0 publish is not the end. Put a reminder:
- 30 days: tool names still accurate?
- 90 days: free-tier still free without card?
- Seasonally: safety examples still responsible?
- When mail arrives from readers: log corrections publicly if the error mattered
A living hive edits. A brochure site pretends forever.
Mini FAQ add-on
Can I automate fact-checking fully? No. Can I skip disclosure if the AI only fixed commas? Usually a light note is enough; be consistent. Can I use AI to write apology letters after a mistake? Yes — then send as yourself. Can Apiary staff see my local Ollama chats? Not by design of local inference; protect your device. Is longer always better? No. This wave's floor exists because these topics need steps and guardrails — not because fluff is virtue.