20 AI Questions to Ask at the Dinner Table (Family Conversation Starters)

Here's the thing: these aren't prompts you type into a chatbot. There's no screen involved at all. These are questions you ask your kid, out loud, over dinner — about AI, robots, and what a future with smart machines might look like.

Kids are already forming opinions about AI from school, YouTube, and friends. Dinner is a low-pressure place to find out what they actually think, correct a misconception or two, and maybe learn something surprising about how your child sees the world. No app required — just you, your kid, and a fork.

📋 Get the Free Prompt Pack — 150 Safe ChatGPT Prompts by Age

Below are 20 questions, grouped by age band since the right question for a 7-year-old ("would you want a robot friend?") is a very different question for a 14-year-old ("should companies have to say when you're talking to a bot instead of a person?"). Pick two or three per meal — you don't need to run through the whole list in one sitting, and honestly you shouldn't. Let one question breathe.

Questions for Younger Kids (Ages 6-8)

Little kids think about AI in terms of magic, robots, and helpers. Keep it concrete and playful.

1. "If a robot could do one chore for you every day, which chore would you pick — and which one would you keep doing yourself?" Good for surfacing what they actually dislike doing (and sometimes the answer is "I'd still want to walk the dog myself," which tells you something nice.)

2. "If you could build a robot friend, what would it look like, and what would it be good at?" This is basically a design prompt in disguise — kids this age love describing imaginary creatures.

3. "Do you think a computer can ever really 'think,' or is it just following instructions really fast?" Simple enough for a 6-year-old to have an opinion, and it opens the door to explaining how AI actually works without getting technical.

4. "If a smart speaker or AI could answer any question in the world, what's the first thing you'd ask it?" Fun to hear whether they ask something silly, something curious, or something surprisingly deep.

5. "Should a robot ever get to say 'no' to a job it doesn't want to do?" Kids this age have strong instincts about fairness — this question taps right into that.

6. "If a robot made a mistake, whose fault is it — the robot's, or the person who built it?" Introduces the idea of responsibility in a way a first-grader can chew on.

7. "What's something you wish an AI could help you build or draw?" A great lead-in if you want to actually try it together afterward.

Questions for Middle Kids (Ages 9-11)

This age group starts weighing trade-offs — fairness, effort, and "is this cheating?" become interesting to them.

8. "If a robot could do your homework perfectly, should it? Why or why not?" This one almost always produces a surprisingly nuanced answer — kids this age usually land on "it depends what the homework is for."

9. "What's something AI is good at that people are bad at — and something people are good at that AI is bad at?" Helps them start building a mental map instead of thinking of AI as either magic or a threat.

10. "If you found out a story, poem, or picture was made by AI instead of a person, would you feel differently about it? Why?" A genuinely interesting question adults don't agree on either — let them sit with it.

Questions for Older Kids (Ages 12-15)

Teens can handle — and usually enjoy — questions with real tension and no clean answer.

15. "Should companies have to tell you when you're talking to an AI instead of a real person? Why does that matter, or not?" A genuinely live policy debate — there's no "correct" answer, which makes it good dinner conversation.

16. "If AI can write an essay as well as you can, what's the point of learning to write one yourself?" Don't rescue this question with an immediate answer — let them wrestle with it first.

17. "Do you think AI will take away jobs, create new ones, or both? Which jobs do you think are safest?" A good one to revisit every six months and compare answers as they get older.

18. "If an AI could predict what you'd want to buy, watch, or do next — is that helpful, or a little creepy?" Opens up a real conversation about data and privacy without lecturing.

19. "Should there be things AI is never allowed to do, no matter how useful it might be?" Great for hearing where they draw ethical lines.

20. "If you were in charge of a big AI company, what's one rule you'd make sure your product always followed?" Turns them from critic into decision-maker for a minute — often reveals what they actually worry about.

How to Actually Use These at the Table

A few things make these land better:

• Ask, then go quiet. The instinct is to fill silence with your own opinion. Don't — let the question sit for a few seconds.
• Follow up with "why?" almost every time. The first answer is rarely the interesting one.
• Don't correct them mid-answer. If they say something factually off about how AI works, let them finish, then gently add context ("that's part of it — here's the other part").
• It's fine to disagree with each other at the table. Modeling a respectful disagreement about AI is more useful than modeling agreement.
• Bring it back to real tools sometimes. If a question sparks curiosity ("wait, can AI actually do that?"), that's a natural moment to sit down together afterward and try it — with you nearby, especially for younger kids, since AI chatbots can produce outputs that need a second look before a child acts on them.

If dinner-table talk turns into "okay, can we actually try that," our guide to kids' ChatGPT prompts by age is a good next stop — it has age-appropriate starting prompts organized the same way these questions are.

📋 Get the Free Prompt Pack — 150 Safe ChatGPT Prompts by Age

FAQ

Do I need to know a lot about AI to ask these questions?

No. Most of these questions work better when you're genuinely curious alongside your kid rather than delivering a lecture. If they ask something you don't know, "let's find out together" is a perfectly good answer.

What if my kid says something that shows a misconception about AI?

Let them finish their thought first, then add context gently — something like "that's partly true, and here's the other part." Dinner conversations work best when kids feel safe thinking out loud, even if the first draft of their opinion isn't fully accurate yet.

My kid isn't interested in "serious" topics at dinner. Any tips?

Start with the sillier, more concrete questions (like #1, #2, or #4) rather than the policy-style ones. Once they're engaged, you can work up to bigger questions over several dinners rather than one sitting.

Should younger siblings hear the older-kid questions too?

It's fine if they're in the room — younger kids often surprise you by chiming in with their own take. Just don't expect a 7-year-old to engage with question 15 the way a 14-year-old would; let each kid answer at their own level.

Further reading

If tonight's conversation turns into "let's actually try this," start with kids' ChatGPT prompts by age for age-matched prompts to try together.