150 Questions Worth Asking Your AI

You probably use AI for a handful of things. Draft an email, settle a quick fact, rescue a recipe. Useful, and a small corner of what it can actually do.

The widest survey of how people use these tools, Marc Zao-Sanders’ 2025 report on the top hundred uses, put therapy and companionship at number one. Organizing a life came second, finding a sense of purpose third. Most of the high-value uses are personal, and most people have never tried them.

So here is a map across the parts of your life that usually go unasked. Eleven areas, 150 questions, each one a real prompt you can copy, fill in the [brackets], and paste into ChatGPT, Claude, or Gemini today. Take the ones that fit. Skip the rest. None of them need any setup beyond the chat box you already have open.

One tip that makes all of them work better: be specific. A closed, concrete question (“I wake at 3am four nights a week, what are the likely causes”) gets a sharper answer than a vague one (“how do I sleep better?”), and in health, where it has been measured, the specific kind also comes back safer, problematic roughly 7 percent of the time against 32 percent for the open-ended kind (a secondary source, so read it as a directional figure). That is why every question below names a real situation instead of asking in general.

And a frame for the heavier ones: some of these are for thinking with, not for getting a verdict. A few touch things where a wrong specific costs you, a dose, a dollar figure, a legal deadline. That is never a reason to stay quiet. Ask, then verify the specifics with a person or the source. Where a section needs that reminder, it gets one line.

Show it, do not type it. All three big apps can see photos now, so a lot of these are faster if you snap a picture instead of describing it: the fridge, the bill, the lab printout, the form, the plant. Where a photo clearly wins, the question says so. (Photographing a printout or a label is just reading text, which is fine. None of these apps read actual medical scans, so the health photo questions stay on the printout, never the X-ray.) If you have never done it, Stuck? Ask It, or Show It is the deeper how-to.

Memory matters. The questions that read you back (“based on what you know about me,” your blind spots, your real values) only work well once you have chatted with one app for a while and turned its memory on. A brand-new chat has nothing to go on.

A word on privacy. For the sensitive stuff, health, money, anything personal, these chats are not private by default, and your inputs may be used to train the model unless you opt out. Do not paste full account numbers or other people’s private details. AI and your data: a three-question privacy check is a one-minute version of what to toggle.

Free PDF · all 150 The whole list, by your keyboard Every question here, grouped by the same eleven sections, sized to print. Keep it nearby and fill in the brackets as you go. Download cheat sheet ↓

Your mind

The most-reported use of AI is also the one that needs the steadiest hand. Every question here is for thinking with, not handing over: journaling, reframing, sorting a tangle of feelings back into named strands. A smarter rubber duck, not a therapist. There genuinely can be benefit, and there is also, in the words of one Harvard psychiatrist, “a subtype of people who clearly should not be using AI chatbots.” For a real crisis, go to a person or a hotline, not a chat window. With that said, these are good for the everyday “I cannot see this clearly” moments.

1. It is late and I cannot stop thinking [the specific thought, like “that I am going to get fired Monday”]. Do not reassure me and do not tell me to relax. Walk me through it: the realistic worst case, how likely it actually is, and one small thing I could do tomorrow morning if it helps.

2. I made [mistake] and I have been replaying it for hours. Give me three other ways to look at what happened, including the most forgiving one a good friend would offer, without pretending it was not a mistake.

3. I have a feeling I cannot name. It has been sitting with me since [whatever happened], it is not quite [sad / angry / anxious], and I keep wanting to [behavior]. Ask me a few questions one at a time, then offer me three or four more precise words for what this might actually be, and tell me why each one fits.

4. Mirror my self-talk back to me. Here is roughly how I have been talking to myself about [the situation], word for word: [paste or describe the things you say in your head]. Read it back as if a friend were saying it to me, then tell me honestly whether it is fair, and where it is harsher or more catastrophic than the facts support.

5. I am spiraling about [situation] and going in circles. Reflect my own thinking back to me in plain language so I can see the loop I am stuck in, then ask me what would have to be true for me to set it down for tonight.

6. Today was rough and I do not even want to explain the whole thing. Interview me about it, short questions one at a time, and at the end tell me back what actually seems to be bothering me most, in case it is not the thing I think it is.

7. I keep avoiding [the specific thing]. I am not asking you to push me to do it. Help me figure out what I am actually protecting myself from by not doing it, what the fear underneath is, and whether it is the kind that shrinks when you look at it or the kind that is pointing at something real.

8. I get a wave of dread every [Sunday night / morning before work / time]. It is predictable but I have never looked at it straight. Walk me through it: what is the story my brain tells in that moment, how much of it tends to come true, and one small thing that might take the edge off next time.

9. I have [the hard thing: a presentation, a confrontation, a first day] in a few hours and I am in my head about it. Remind me of what I have actually got going for me here, and give me one steadying thing to tell myself walking in. No fake hype.

10. Give me the honest pep talk, not the cheerleader one. I am doubting whether I can [goal]. Acknowledge what is genuinely hard about it, then make the real case for why it is still worth trying.

11. I want to cut down on [doomscrolling at night]. Do not shame me. Help me figure out what I am actually getting from it, then suggest two swaps that scratch the same itch in a better way.

12. I have a hundred things in my head and I cannot start any of them. Let me brain-dump everything right now, then help me pick the single next action, small enough to do in ten minutes, and tell me to ignore the rest for now.

13. I lost [person] and there are things I never got to say. Help me write a letter to them I am never going to send, just to get it out of me. Ask me what I would most want them to know before we start.

14. Today is hard because of [loss] and my instinct is to stay busy and push it down. Instead, help me sit with it for a few minutes. Ask me what I am missing most about them today, and just let me say it.

Those last two stop at your words. None of this asks the AI to be or speak as the person you lost, on purpose. And if grief or anything else feels like more than you can carry, that is a moment for a person, not a chatbot.

Money and the big decisions

A man fed an itemized hospital bill into AI and it flagged duplicate charges, phantom “inpatient” billing, and supply costs inflated far above normal rates. The bill went from $195,628 to $33,000. One detail does the honest work, though: the AI surfaced the leads, and a human verified each one and wrote the dispute letter. That is the shape of every money question here. AI reads the dense thing and finds the threads. You, or a professional, confirm the dollar figures, the deadlines, and the statutes.

15. I am pasting an itemized hospital bill below, with the billing codes. Go through it line by line and flag anything worth questioning: duplicate charges, services billed but not received, a procedure marked “inpatient” when I was never admitted, or supply costs that look far above normal rates. For each flag, tell me what to ask the billing office. [paste itemized bill]

16. Here is a photo of my [restaurant receipt / repair invoice / itemized quote]: [attach photo]. Read it back to me line by line, flag anything that looks like it was added without my okay or charged twice, total it yourself, and tell me which single line I would be most justified in questioning.

17. I am about to sign this [lease / contract / offer letter / freelance agreement]. Explain it in plain language, then list the clauses most worth paying attention to before I sign: anything unusual, anything that favors the other side, and any blanks or vague terms I should get pinned down in writing. [paste document]

18. Read these terms of service and tell me in plain English what I am actually agreeing to, especially what data they collect, what they can do with it, whether I am giving up the right to sue or join a class action, and how I would cancel. [paste or describe the terms]

19. Here is a photo of every subscription on my [bank / card] statement this month: [attach photo, or paste the recurring lines]. Sort them into “use it and worth it,” “use it but overpaying,” and “forgot this was even charging me.” For the last group, tell me how to cancel each one.

20. I am standing in the store deciding between these two: [photo of both products with their shelf price tags, or paste size and price for each]. Which is the better deal per [unit / ounce / count], are there hidden trade-offs I am not seeing, and is the bigger one actually cheaper or just counting on me to assume so?

21. Do not answer this for me yet. I want you to interview me first. I am trying to decide [the decision]. Ask me one question at a time until you genuinely understand my situation, what I care about, and what I am afraid of. Then reflect the real tradeoffs back to me. Do not tell me what to do.

22. I am weighing [option A] against [option B]. Here are my actual constraints: [budget, timeline, location, what I cannot compromise on, who else this affects]. Walk me through how each option holds up against each constraint, and tell me where I am likely underweighting something.

23. I want to negotiate [my rent / this quote / a bill] and I have never done it and I am nervous. Here is the situation and the actual number: [paste]. Give me the opening line, my one or two strongest bargaining points, the number to anchor on, and what to say when they push back, short enough that I can remember it on the call.

24. I am meeting with a [tax preparer / financial advisor / mortgage broker] next week about [topic]. Explain the basics well enough that I can follow the conversation, then give me the smart questions to ask so I do not get steamrolled or sold something I do not need.

25. My open-enrollment health plan options are below. Given that I [take this medication / have this condition / mostly stay healthy], help me understand the real cost difference past the monthly premium: deductible, out-of-pocket max, and how my situation actually hits them. [paste plan summaries]

26. I might leave a stable job for [riskier thing]. Help me name what I would actually be trading, money, security, identity, time, the people I work with, and pressure-test the story I am telling myself about why this is the right move.

27. A salesperson is pushing me toward [an extended warranty / a timeshare / whole-life insurance / a financing offer]. Explain how this kind of product actually works, who it is usually good or bad for, and the questions that expose whether it is a good deal for me specifically.

28. Translate this denial into plain English and tell me my options. My [insurance claim / loan / appeal] was denied, here is the letter. What does it actually say, is there a deadline to respond, and what would a strong appeal need to address? [paste the letter]

29. Before I commit, argue me out of it. I have basically decided to [big financial or life decision]. Make the strongest honest case against it, the risks I am glossing over and the questions I have not asked, so I am choosing with my eyes open.

A denial letter has a real deadline and a contract has real terms. Confirm those against the document itself before you act. AI can misread a clause, and missing a real deadline is expensive.

Relationships

The best questions here are the non-obvious ones: rehearse the hard talk against the version of someone who is having a bad day, decode a flat text before you fire back at a mood that is not there, feed it both sides and ask what each of you is missing. As the psychologist Hal Shorey, who uses it this way, notes, the tool excels at wording messages. It is not bound by any privacy rule, it has no stake in being fair to the person who is not in the room, and its read flips with how you frame the question, so it finds the blind spot. It does not settle the score.

30. I have a hard conversation coming up with my [boss / mother / partner]. I want to practice. You play them: here is what they are like, [a few sentences on their personality, what they tend to say, what sets them off]. I will start, you respond as they actually would, including pushing back the way they really do. Do not go easy on me.

31. Here is a text I just got from [who]: “[paste it exactly].” I cannot tell if they are annoyed, joking, or completely neutral and I am reading into it. What are the possible things they could mean, and what in the wording points to each one?

32. Here is a back-and-forth from a group chat that is getting tense: [paste the last 10 to 15 messages, names changed]. I cannot tell what is actually going on under the surface. Who seems upset and who does not, what is the unspoken thing the argument is really about, and is there a message I could send that would lower the temperature instead of adding to it?

33. I need to tell [who] [the hard thing, like “I cannot make it to the wedding”]. Word it three ways: one warm and soft, one direct and brief, one in the middle. Keep it honest in all three. Do not make it gushy or fake.

34. I want to set a boundary with [who] about [the thing]. Help me word it so it is clear and kind, firm about the boundary but not an attack on them. Give me a gentle version and a slightly firmer one for if the gentle one gets ignored.

35. I owe [who] an apology for [what happened]. Help me write one that actually takes responsibility: no “I am sorry you felt that way,” no excuses smuggled in, no over-explaining. Then flag any line that still sounds defensive.

36. I want to write a real thank-you to [who] for [what they did / who they have been], not a generic one. Interview me first: ask me for two or three specific moments with them, then help me turn those into something that sounds like me and would actually mean something to them.

37. I am stuck in a fight with [who]. Here is my side: [paste]. Here is their side, in their own words as best I can give it: [paste]. Do not tell me who is right. Tell me what each of us seems to be missing about the other, and what each of us actually wants underneath the argument.

38. [Who] and I have the same fight over and over about [the recurring thing]. Here is how it usually goes: [describe the loop]. What is the pattern underneath it, what is each of us probably reacting to, and is there a different way I could open it next time so it does not go down the same track?

39. I reacted way bigger than the situation deserved when [who] did [the thing], and I do not fully understand why. Here is what happened: [describe it]. Help me trace it back, what nerve it probably hit, what older thing it might be echoing, and what I would want to understand about myself before I bring it up with them.

40. I had a conversation with [who] that did not go how I wanted, and I keep replaying it. Here is roughly what was said: [paste or describe]. Help me figure out what I actually wish I had said, and whether it is worth following up now, and if so, how I would word it.

41. I tend to spin out when there is tension with [who]. Here is the situation: [describe]. Help me get my head straight first: what is the most likely read here versus the worst-case story I have built, and what would a calmer version of me do next?

42. There is a person in public life I find genuinely baffling, [name them, or describe them]. Without telling me they are secretly right, reconstruct how the world might look from inside their head so that what they do feels obvious, even necessary, to them. What are they seeing or fearing that I am not?

43. Here is a message I drafted to [who] while I was upset: [paste]. Do not soften my point, keep what I actually mean. But flag anything likely to land as a jab or escalate, and suggest a cleaner way to say that part.

A line worth keeping in view: the AI only knows what you typed, and only ever hears your side. Use it to find the words and the blind spot. For anything consequential, talk it through with the real person, and with someone you trust if it is a big call.

Your health and body

This is the section where being specific does double duty. Every question is framed to understand, get a second perspective, or walk in better prepared, never “diagnose me” or “decide my treatment.” That is partly a safety thing: closed, specific health questions came back problematic about 7 percent of the time, against 32 percent for open-ended ones. More than 230 million people a week now ask ChatGPT health questions, so the careful way is worth knowing. Emergencies go to a human. Anything that drives treatment, a dose, an interaction, a diagnosis, gets confirmed with a clinician or the source. And the photo questions below are for a picture of the printout or the label, never an actual scan.

44. Here are my blood test results: [paste the values and the lab’s reference ranges, or attach a photo of the printout]. Explain each one in plain English, what it measures, whether mine is inside the normal range, and which ones, if any, are worth asking my doctor about.

45. This is the findings section of my [imaging report / pathology report]: [paste it]. I have no medical background. Rewrite it in plain language, define every term, and tell me what questions it would be reasonable to ask the doctor who ordered it.

46. I am not asking for a diagnosis, I want a second perspective to prepare for my appointment. My symptoms are: [be specific: what, where, how long, how severe, what makes it better or worse]. What are the common explanations a clinician would consider, what would they likely ask me, and what would make this urgent enough to be seen sooner?

47. Here is a photo of [a rash / a bruise that will not fade / a swollen spot]: [attach photo]. It has been [how long], it [itches / hurts / does nothing], and it [is / is not] spreading. I am not asking for a diagnosis, give me the common explanations, what would make this worth seeing someone about soon, and what would make it an emergency.

48. I have a 15-minute appointment about [specific concern] and I always forget half my questions. Give me a short, prioritized list to ask, and the three pieces of my history the doctor will most want from me.

49. I have felt dismissed before and I get anxious advocating for myself. My situation is [describe it]. Help me word my main concern clearly and calmly in two or three sentences I can actually say out loud, and suggest how to respond if I am brushed off.

50. I keep waking around 3am and cannot fall back asleep, three or four nights a week, for about [time period]. Walk me through the most common causes, what is reasonable to try first, and what would be worth mentioning to a doctor.

51. For the last two weeks I have logged my [headaches / energy crashes / joint pain]: [paste your notes: times, food, sleep, stress, anything you noticed]. Look for patterns or possible triggers I might have missed, and tell me what a clinician would want me to track next.

52. Here is what I ate yesterday: [list every meal and snack]. Estimate the rough calories and the protein, carbs, and fat breakdown, point out what the day was light or heavy on, and suggest one realistic swap, not a whole diet overhaul.

53. Here is a photo of both sides of a pill I found loose in [a drawer / my parent’s old kit], including any imprint, color, and shape: [attach photo]. What are the possible matches, and what is the single safest thing to do before anyone takes it?

54. Here is a photo of the full ingredient and dosage panel on this [supplement / cold medicine / vitamin]: [attach photo]. I already take [list your meds and other supplements]. In plain English, what is actually in this, what is it doing, and is there anything here worth running past a pharmacist before I combine it with what I am on?

55. I want to upload a short clip of my [squat / deadlift / kettlebell swing] to Gemini and ask what it notices about my form. (Video upload is a Gemini thing; ChatGPT and Claude do not take video, and the free tier caps at about five minutes.) Treat it as a second set of eyes, not a coach: it is much better at confirming a rep that looks roughly right than at catching a genuinely bad one, so use it as a prompt to double-check yourself, not a verdict.

56. I have been prescribed [medication name and dose]. In plain English: what is it for, how does it generally work, what are the common side effects, and what are the most important questions to ask my pharmacist when I pick it up?

57. Explain what is happening inside my body right now [at X weeks pregnant / during a migraine / when I get the post-lunch slump], and if it helps, draw me a simple diagram.

58. Here is a photo of how I sit at my desk all day, from the side: [attach photo]. My [neck / lower back / wrist] aches by the afternoon. Point out what is likely working against me in this setup, and give me three changes I can make today with what is already in the room.

A reminder that carries the section: a result, a label, or a photo can read scarier or milder out of context than it is. Use these to understand and to prepare your questions. The clinician who ordered the test interprets it, the pharmacist is the free expert who can check a real interaction, and anything spreading, darkening, or paired with feeling unwell gets a real visit.

Learning something, or just being curious

The core of this domain is not “summarize this for me.” It is the follow-up: make it explain at the right level, find the gap in what you understand, then go one rung deeper. The most-loved example of the year was a small one. Someone asked ChatGPT to explain their job (Senior Data Analytics Consultant) to a five-year-old, got back “someone who counts things for other people because those people are too busy,” and reported never feeling so professionally attacked by such an accurate description.

59. Explain [my exact job: paste your real title and a one-line description] in plain language, the way you would describe it to a smart friend who has never worked in my field. Then tell me one thing about it I probably take for granted that an outsider would find strange.

60. Here is a photo of [the page / diagram / chart / handwritten note] I cannot make sense of: [attach photo]. Walk me through it like I am seeing it for the first time, what each part is, what it is actually telling me, and the one thing it most wants me to understand. If my handwriting or the image is unclear anywhere, tell me where so I can retype it.

61. I am reading this [news story / explainer] but I do not have the background to follow it: [paste the text or the headline and key paragraphs]. Explain it as if I missed the last five years of context on this topic. What is the backstory I would need to actually understand why it matters?

62. Quiz me on [topic] one question at a time. Start easy, and every time I get one right, make the next a little harder. If I get one wrong, explain why, then stay at that level until I am solid. Keep going until I am clearly out of my depth.

63. I want to really understand [how mortgages work / what a tariff actually does / how my immune system fights a cold]. What is the prerequisite I am probably missing that would make the rest click? Ask me one or two quick questions first to find where my gap actually is.

64. I think I understand [topic], but I want to find the holes. Ask ME questions about it, one at a time, starting basic and getting more probing, and do not tell me the answers. When you spot something I clearly do not understand, stop and tell me what the gap is.

65. Give me an analogy for [the hard thing I am trying to grasp] built entirely on something I already understand well, like [a thing from my job, hobby, or daily life]. Then tell me where the analogy breaks down, so I do not over-trust it.

66. Look at [this thing: paste a photo, a chart, a paragraph of writing, a chess position, a recipe]. What would an expert immediately notice that a beginner like me would walk right past? Point to the specific details and why they matter.

67. I am looking at a menu, sign, or label in [language] and I am a bit lost. Here is a photo: [attach photo]. Translate it, but also tell me what you would actually order or do here if you were me and [I am vegetarian / I hate cilantro / I am trying to find the bathroom], not just a literal word-for-word.

68. I just read [paste it or describe it]. What should I be skeptical of here? Point out claims that are weak, unsupported, cherry-picked, or that the author has an incentive to push, and tell me what would change my mind in either direction.

69. Something just broke or surprised me: [the boiler made a noise, an error code came up, a plant on my hike]. Walk me through what is likely going on AND explain the underlying thing well enough that I would understand it next time, not just fix it once.

70. I keep hearing [a jargon term: “amortization,” “the Maillard reaction”] and nodding along without really knowing it. Define it in one plain sentence, give me a concrete everyday example, then show me how it is used in a real sentence so I would recognize it next time.

71. Introduce me to a thinker I have probably never seriously read but would find genuinely interesting, given that I care about [the things you know I care about / these topics: ___]. Tell me their one big idea in plain language, the thing that makes people argue about them, and one question of mine they would answer completely differently than I do.

72. Show me an entire way of seeing the world that I have probably never encountered, not a famous one, [something like animism, a school of thought from outside the West, a fringe-but-serious idea]. Lay out how it carves up reality differently than I do, where it is genuinely insightful, and where I would push back. Then ask me which part snagged me.

73. Here is a photo of a [parking sign / posted notice / set of rules] I cannot decode: [attach photo]. It is [Tuesday, 4pm, this neighborhood]. In one plain sentence: can I park here right now, for how long, and what is the thing most likely to get me ticketed?

When the topic is recent or contested, treat the explanation as orientation and check the specifics against the real reporting. AI can be confidently out of date.

Parenting and family

The most useful thing AI does here is translation: you know the answer to the hard question, you just do not have the words for a specific five-year-old standing right there. After that it is the constraints, a rainy Saturday and one weird obsession and no craft supplies, that turn a generic activity list into a usable one.

74. Help me explain [where babies come from / why people look different / what “dead” means] to my [5]-year-old at a level they will actually understand. Keep it honest but age-appropriate, and give me the words to use if they ask a follow-up I will not expect.

75. My [6]-year-old just asked why [Grandpa died / their friend has two moms / some kids do not have enough food]. They are standing right here. Give me three or four short, calm, true sentences I can say right now, and tell me what NOT to say.

76. Make up a bedtime story starring [my kid’s name], age [4], who loves [dinosaurs and the color purple], with a gentle lesson about [sharing with their little brother]. Keep it about five minutes to read aloud, and end on something calm and sleepy.

77. Turn [my 7-year-old’s name] into the hero of a six-chapter bedtime adventure, one short chapter a night. They love [space and their dog Biscuit]. Leave each chapter on a tiny cliffhanger, and ask me one question at the end of each so they help decide what happens next.

78. It is a rainy Saturday and my [4]-year-old is obsessed with [garbage trucks]. Give me six indoor activities I can set up in five minutes with stuff I probably already have, no screens, that lean into the [garbage truck] thing.

79. Gift ideas for a [picky 11-year-old] who [has read all of Harry Potter, loves chess, and is into science museums], for [under $30]. Skip the obvious stuff they probably already have, and tell me why each pick fits.

80. I need to bring up [vaping / a bad grade / something I saw on their phone] with my [15]-year-old without them shutting down or feeling attacked. Help me figure out how to open it, what tone to aim for, and a couple of questions that invite them to talk instead of defend.

81. Play my [16]-year-old in a tough conversation about [their curfew]. Take their side and push back like a real teenager would. I will practice staying calm and not lecturing. After a few rounds, tell me where I got defensive or shut them down.

82. Is it developmentally normal for a [3]-year-old to [suddenly start hitting / refuse to sleep alone / melt down over the wrong cup]? Walk me through what is likely going on, what is usually a phase, and a few calm ways to respond, and flag anything I should actually mention to the pediatrician.

83. Turn [brushing teeth / cleaning up toys / getting dressed] into a game for my [4]-year-old who [loves pretending to be a superhero]. Give me three quick versions I can rotate so it does not get stale, and keep it fast, we are always running late.

Great for context and for calming your own 2am panic, but it does not know your kid. Anything about their development, behavior, or health that worries you gets confirmed with the pediatrician, who can actually see them.

The everyday stuff

The closest-to-home use, and the one people are most surprised AI is good at. Ethan Mollick keeps a running list of his own: identifying plants on a hike, getting cooking tips with his hands covered in flour, building a Wisconsin travel guide focused entirely on cheese. The trick in every one is the constraint: tell it what is actually in your fridge, who is actually on the trip, what the noise actually sounds like. And when describing is a hassle, show it the thing instead.

84. Here is exactly what is in my fridge and pantry right now [eggs, half a bag of spinach, feta, a lemon, leftover rice, two chicken thighs], or a photo of the open shelves. What can I make for dinner tonight without a store run? Give me two or three options, fastest first.

85. Plan me a realistic [4]-day trip to [Lisbon] for [two adults in their sixties who walk slowly and hate crowds]. Sane pacing, one or two things a day with downtime, not a tourist firehose. Group things by neighborhood so we are not crisscrossing the city.

86. Identify this plant from a photo and tell me whether it is safe for [my dog / a curious toddler] to be around. [attach photo, or describe it: waxy oval leaves, small white flowers, grows low along a fence]

87. Here is a photo of [the plant that is struggling], leaves and pot included: [attach photo]. It [droops / has yellow leaves / has brown tips] and I [water it weekly / keep it by a north window]. What is most likely wrong, what should I change this week, and is this one savable or am I better off starting over?

88. I keep hearing a [rhythmic clicking / low hum / grinding] noise from [my fridge / under the car when I brake / the wall behind the bathroom]. It happens when [it has been running a while / I turn left]. What are the likely causes, ranked, and which ones are fine to ignore versus call someone now?

89. I got this error message: “[paste the exact error].” I was trying to [print a document / install an app / log into my bank]. Explain in plain English what it actually means, then walk me through fixing it step by step. Assume I am not technical.

90. Walk me through troubleshooting my [dishwasher that will not drain / router with no internet / thermostat that will not turn on], one step at a time. Ask me what I am seeing after each step before giving me the next one. Do not dump the whole list at once.

91. I am pasting a [letter from the city / official form / notice from my insurer] below, or a photo of it. Tell me in plain English what it is actually asking me to do, whether there is a deadline, and what happens if I ignore it. [paste text or attach photo]

92. Here is a photo of [the room / corner / wall] I want to rearrange: [attach photo]. I want it to feel [calmer / bigger / more useful for working] and I am not buying much. Tell me what to move where using only what is already in the shot, and what one cheap thing, if any, would actually make the difference.

93. Here is a photo of my closet or a pile of clothes, and the event is [a job interview / a casual dinner / a funeral]: [attach photo]. Put together one or two outfits from what is actually here, tell me why each works for the occasion, and flag anything that would read as too formal or too casual.

94. I am hosting [a 6-year-old’s birthday party for 12 kids] with a budget of [$150] and [a small backyard]. Plan it: a timeline for the day, food, two or three activities, and a shopping list that stays under budget.

95. Give me a low-effort weeknight dinner rotation I can actually stick to, something like Pasta Mondays and Taco Tuesdays, built around [meals I can make in 30 minutes with two kids underfoot]. Make it repeatable so I stop deciding every night.

96. I am packing for [a 5-day trip to a cold, rainy city] and I always overpack. Give me a tight packing list for that weather and length, organized by category, assuming I will do laundry once. Push back if I am bringing too much.

What to watch, read, and listen to next

Recommendation engines match on genre. What works here is making AI match on the thread in your taste, which is the thing a menu of “because you watched” rows can never find.

97. I loved [three to five titles: movies, shows, books, or albums]. Recommend things I have probably never heard of, and tell me what they have in common. What is the actual thread running through my taste?

98. Given what I love ([list three or four titles]), what is the older, foundational work that the things I love were all quietly influenced by, the one I have somehow never gone back to? Tell me what to look for in it so I can see the through-line to the stuff I already love.

99. I am trying to name a [movie / show / book / song] I half-remember. Here is everything I have got: [a scene, a feeling, roughly when I saw it, an actor, a plot fragment, a line]. What is it?

100. Name this song. The lyrics I remember are roughly “[mangled lyrics, even if the words are wrong].” It sounded kind of [genre or mood], probably from around [decade], and the [chorus / hook] went something like [hum it in words].

101. Break me out of my rut. I keep watching [genre or type], like [name three recent examples]. Recommend something genuinely good that is unlike what I normally pick, and tell me specifically why I might still love it.

102. I want to get into [director / author / band / a long series]. Build me a watch, read, or listen ORDER: where to start, what to skip on a first pass, and what to save for when I am hooked. Tell me why for each.

103. I have got [2 hours] tonight and I feel [wrung out / restless / cozy / wired]. I do not want to think hard about what to put on. What should I watch, and why does it fit how I feel right now?

104. There is an art form I just do not get and quietly think is for show-offs: [opera / abstract painting / poetry / jazz / a sport]. Do not tell me I am wrong. Show me what its devotees are actually experiencing that I am missing, give me one specific thing to look or listen for, and one piece to try with that in mind.

105. I have a very specific itch: [a heist where the plan goes wrong in the first 20 minutes / a cozy mystery set somewhere cold / sci-fi that is hopeful instead of bleak]. Find me [movies / shows / books] that scratch exactly this.

106. I bounced off [the acclaimed thing everyone loves] and felt like I was missing something. Explain what people see in it without telling me I am wrong for not connecting with it, then tell me whether it is worth a second try for someone like me.

107. I want to understand [film noir / city pop / a music genre I know nothing about]. Give me a starter flight of five to experience in order, what to notice in each, and how they connect, so I come out understanding it instead of just having sampled it.

One honest note: AI will sometimes name a title that sounds real and is not. Before you go hunting for a deep cut it recommended, ask it to confirm the thing exists, or just search the name.

Making things

The thread across this whole section is the same: AI is a yes-machine by default, so the good prompts make it switch sides, push past the safe first idea, or refuse to solve the thing for you. The best creative use is often the question it asks back, not the answer it gives.

108. Give me 20 ideas for [a name for my pottery side-business / a theme for my daughter’s birthday party / what to do with the empty corner of my living room]. Then take the five weirdest ones and push each one much further than is reasonable.

109. Be my brainstorm partner. Here is my half-formed idea: [a podcast where I interview people about the worst job they ever had]. First build on it, give me the best version and three directions it could go. Then switch sides and poke holes: what would make this fall flat, and what is the obvious trap?

110. Run a tabletop adventure for me as the narrator. The setting is [a haunted lighthouse on a dying coast]. I am playing [a retired smuggler with one good eye and a debt to pay]. Describe each scene, then stop and ask me what I do, and let MY choices drive the story. Do not decide for me.

111. I am writing [a mystery novel set in a 1920s seaside hotel]. Help me develop the world: ask me eight questions about the place, the era, and the secrets it is hiding, one at a time, and build on each answer. Do not write the story yet, just deepen the world with me.

112. I am stuck on [the second act of my screenplay] and I have been stuck for weeks. Do not give me ideas or solutions. Instead, ask me what I am avoiding: the scene or decision I keep skipping past because it scares me or I do not know how to write it.

113. Take this thing, [my neighborhood coffee shop], and reimagine it as if it were designed by [a heavy-metal band / a Studio Ghibli film / the people who make luxury watches]. Walk me through the menu, the vibe, the name, and one detail nobody would expect.

114. What is the most obvious, predictable, seen-it-a-hundred-times version of [my idea for a coffee-shop loyalty app], so I can make sure I am NOT building that? List the clichés and the moves everyone makes, then suggest what the non-obvious version would do differently.

115. I have a chapter of [my novel] right here. Do not rewrite it. Read it as a sharp, generous editor would and tell me: where did your attention drift, where did you not believe a character, and what is the one change that would do the most good? [paste chapter]

116. I want to make [a short comic / a zine / a tiny video game] but I have never done it before. Do not plan the whole thing. Just tell me the smallest possible version I could finish this weekend, and the very first concrete step I would take to start.

117. I keep making the same kind of thing, [all my songs end up as slow sad ballads]. Show me my pattern back to me, name what I reliably reach for and what I reliably avoid, then give me one constraint that would force me out of my rut for the next piece.

Knowing yourself

Most of these build their own context: you answer a few questions, or paste in the raw material, and the AI works from what you just gave it. A few read you more sharply if you have chatted with one app for a while with memory on, and those carry a shortcut line. Either way, they are pattern-matching on your own words, not a window into your soul. The author of the most-shared set of these prompts, Jose Casanova, says it plainly: they are “for fun, not professional advice.” Genuinely fun and genuinely useful, just not an oracle.

118. Interview me with ten questions about my life right now, one at a time, and wait for each answer before asking the next. Do not tell me why you are asking. After the tenth, reflect back the patterns you notice about me.

119. Interview me about how I handle [an area I want to look at: my work, my money, my closest relationship], one question at a time, six or seven questions, and do not tell me where you are going. Then name the blind spot you would not have guessed from how I describe myself, and point to the exact answer that gave it away.

     Memory on? Skip the interview and just ask: “Based on what you actually know about me from our chats, what are my blind spots, with specifics from things I have said?”

120. Be my honest coach for a minute, not my cheerleader, and do not flatter me. Here is something I keep meaning to do and keep not doing: [name it]. Tell me the real reason I am avoiding it, then tell me why the excuse I am about to reach for is one I have half-convinced myself is good.

     Memory on? You can drop the setup and ask: “Based on what you know about me, what am I avoiding, and why do I think my excuse for avoiding it is a good one?”

121. Name two things I clearly value that are quietly at war with each other in my life right now. Here is the tension I keep feeling: [describe a situation where you feel pulled both ways]. Tell me which value I have actually been choosing, which one I have been starving, and what that pattern is costing me.

122. Imagine the version of me 30 years from now who already worked through [the situation I am stuck on]. Write me a short letter from them: what they wish I had understood now, and the one thing they would tell me to stop worrying about.

123. I will describe how I actually spent my time and money over the last two weeks, the real version, not the aspirational one. Read it back to me as a list of what I actually value, based on where it went, even if that is uncomfortable.

124. Here are two or three times I can remember being criticized or proven wrong, and how I reacted each time: [describe them]. What is my pattern when my self-image takes a hit, do I dig in, go quiet, overcorrect, laugh it off, and what would a healthier version of that reaction look like for me?

125. Here is my actual calendar from last week: [paste or describe how the days were spent]. Do not judge it. Sort the hours into “lit me up,” “drained me,” and “neutral but necessary,” and ask me a couple of questions first if you need to. Then tell me the one recurring thing I should protect and the one I should question.

126. Here are three or four situations from the last couple of years that felt frustratingly similar [describe them]. What is the recurring pattern I keep walking into, and what is the part of it that is actually me?

127. I will tell you about a decision I am proud of and one I regret: [describe both honestly]. Do not comfort me about the regret. Tell me what the two have in common about how I actually make choices, my real decision-making style rather than the one I would like to claim, and where that style tends to serve me versus burn me.

128. Here are three things I am pretty sure about: [list them]. Pick the one where my confidence is least earned, where I am leaning on a single source, or it is mostly a feeling, or I have never really checked, and tell me why my certainty is running ahead of my evidence.

     Memory on? Ask instead: “Go through the things I have told you I am confident about and find the one where my confidence is least earned, and tell me why.”

129. I want to change [a specific habit] but “just be more disciplined” never works for me. Design me one small, concrete experiment to run for two weeks, something measurable with a clear if-this-then-that trigger, and tell me what result would mean it is working.

130. Here is a situation I am in: [describe it]. What am I pretending not to know here, the thing I probably already know deep down but keep finding reasons to ignore? Name it plainly.

     Memory on? Shorten it to: “Based on what I have told you, what am I pretending not to know about [the situation]?”

131. Here is a story I tell about myself: [the story, like “I am someone who puts family first”]. And here is how my last couple of weeks actually went: [a few honest specifics about where the time and energy went]. Where does the story not match the details, and what might that gap be telling me? Be kind but honest.

     Memory on? Ask directly: “What is a comfortable story I tell about myself that the actual details of my life do not support? Point to specifics.”

132. Give me a kind, specific eulogy for the version of me I am slowly leaving behind, the person I used to be that I am outgrowing. Based on who I was [a few years ago] versus now: [describe the shift]. Honor what that older version got right, and name what I am actually carrying forward.

The big questions

The honest framing for this whole domain comes from the philosopher Harvey Lederman, who wrote about exactly this: you are not here for a verdict on the meaning of life. You are here to ask the question more precisely. Used well, AI is a sparring partner that can make you less sure in a good way, hold a hard question open without rushing to tie a bow on it, and show you the gap between the answer you want and the answer you believe.

133. Steelman the view I disagree with on [open borders / the existence of God / eating meat]. Make the case so well that I understand why a thoughtful, informed person holds it. Assume I already know the weak arguments, give me the strong ones.

134. Here is something I believe pretty strongly: [name it]. Do not argue with me. Reason out loud instead: what would I most likely believe about this if I had been born into [a very different country / a much poorer or richer family / 200 years ago]? Walk me through how a thoughtful version of me in that life would have arrived there honestly.

135. I keep seeing [a group of people: a religion, a movement, a generation, a profession] do something I find baffling: [the thing]. Walk me through their actual reasoning from the inside, charitably, what they are optimizing for, what they are protecting, what the world looks like from where they stand. Assume they are as smart and well-meaning as I am.

136. Explain [a historical belief or practice that seems obviously wrong to us now: ___] from the inside, the way a thoughtful person living then would have defended it. What did it solve for them, what did they know that I do not, and what were they actually afraid of? Do not excuse it, just help me understand how a reasonable person believed it.

137. I am going to argue for [a position I personally disagree with] as if I genuinely held it, and I want you to grade me. After I make my case, tell me honestly: would someone who actually believes this recognize themselves in what I said, or did I strawman them? Where was the real, stronger version different from mine?

138. I believe [X]. Do not argue with me. Just tell me, concretely, what would have to be true, what evidence, what experience, what argument, for me to honestly change my mind. Then tell me whether that thing is even checkable, or whether I have quietly made my belief impossible to disprove.

139. Walk me through the trolley problem one decision at a time. After I answer, change one variable: make it five strangers versus one person I love, then a switch versus pushing someone, then a doctor harvesting one patient to save five. Tell me exactly where my answers stop being consistent and what principle I just quietly abandoned.

140. Describe the experience machine, a tank that gives me any experience I want while I float unaware it is not real. Ask me whether I would plug in for life. Then poke at my reason for saying no, because most people say no and cannot fully explain it. What does my answer reveal about what I actually value besides feeling good?

141. Hand me a thought experiment built to attack a specific belief I hold: [name the belief]. Not a riddle with a clever answer, a real scenario where my own principle, followed honestly, leads somewhere I will not like. Then ask me what I want to do: bite the bullet, find a principled exception, or admit the belief needs work.

142. I want to actually think about [a polarizing issue] instead of just having a side. Lay out the strongest, most honest version of each major position, including ones I would dismiss, as if a smart, decent person held it. Map where they genuinely disagree about facts versus values. Do NOT tell me where to land. At the end, ask me which crux I am least sure about.

143. Here is a situation I am stuck on: [describe it]. Walk me through how a Stoic, a Buddhist, an existentialist, and a utilitarian would each approach it. For each one, give me the core idea, what it would have me do here, and where it might lead me wrong. Then ask me which one I flinched at.

144. I keep wrestling with [a recurring worry, like that I am wasting my one life]. Do not reassure me and do not resolve it. Have a real back-and-forth: ask me what is underneath the worry, push gently on my answers, and help me find a better-shaped version of the question rather than an answer.

145. Let us do an honest back-and-forth about mortality, mine specifically. I am [age] and [whatever is on my mind about it]. Do not comfort me with platitudes and do not pretend to have answers. Ask me what I would actually regret, what “enough time” would even mean to me, and help me think about it more clearly instead of less.

146. Take a belief I hold strongly, [name it], and play the strongest version of my future self who has changed their mind. Tell me what I would say to talk present-me out of it, and what evidence or experience would have done it. What is my belief’s actual weak point?

147. Help me name my own philosophy of life. Ask me a series of questions about what I value, how I make hard calls, and what I think a good life looks like, one at a time, do not explain why, then tell me which established schools of thought I am closest to and where I am a contradiction.

148. Give me a paradox that should genuinely bother me and that I probably have not met, not a riddle with a trick answer but a real one (like the Sorites heap, Newcomb’s problem, or the lottery paradox). Explain why smart people still argue about it, then ask me where my gut wants to escape and why that escape does not fully work.

149. Pretend you are three different philosophers having dinner with me, a Stoic, a hedonist, and a nihilist, and have them argue about [a choice I am facing]. Let them interrupt each other. At the end, ask me whose argument I most wanted to be true, and whose I think is most correct, because they are usually not the same.

150. Let me see a single ordinary day through eyes very different from mine. Pick someone whose life barely overlaps with my own, [a night-shift nurse / a 90-year-old / a refugee / a monk / someone with a disability I do not have], and walk me hour by hour through what they notice, worry about, and take for granted that I never would. Make it specific, not a sympathy tour.

That is the map, all 150. You will not use most of them, and that is fine. The point was never to ask everything. It was to notice that the few questions you already ask are a corner of a much bigger room, and the door to the rest is the same chat box, one specific question at a time.