At 8:15 on a Tuesday morning, two ordinary problems can quietly consume an entire day: figuring out whether a major purchase is genuinely worth the money, and finding a professional way to say no without sounding cold or evasive. AI can help with both - not by making the decision for you, but by organizing the evidence and helping you say what you already mean.

The trick is to use AI as a thinking partner, not an oracle. Ask it to compare, question, summarize, and revise. Do not ask it to “just tell me what to do.” Try at least one before Sunday. That is the whole assignment.

Use Case 1 of 2

Use Case 1: Researching a Major Purchase Decision

At 7:40 p.m. on Wednesday, Linda Morales sat at her kitchen table with three browser tabs open and a headache beginning behind her eyes. Her twelve-year-old car had started making an alarming grinding noise, and the repair estimate was $2,900. A replacement car would mean a monthly payment, insurance changes, and several weekends visiting dealerships. Linda, 58, was perfectly capable of making the decision. What she did not have was a clear way to compare reliability, total cost, safety, resale value, and the practical question she kept postponing: would a newer car actually reduce stress, or simply create a different kind of financial pressure?

This is where AI can be useful. It can turn a vague, emotional purchase into a set of questions and comparisons. It can help you identify costs people commonly forget, such as taxes, maintenance, warranties, fuel, delivery, subscriptions, and financing. It can also point out what still needs independent verification. That last part matters because AI can produce confident but inaccurate information, sometimes presenting invented or outdated details as fact.

Here is the exact prompt to use:

Copy and paste this into ChatGPT or Claude:

> I am researching a major purchase and want help thinking clearly, not a sales pitch. > > The purchase is: [describe the item, model, or service]. > My alternatives are: [list the other options, including keeping what I already own]. > My budget is: [give a total budget or comfortable monthly amount]. > My intended use is: [explain how, where, and how often I will use it]. > The purchase would last approximately: [number of years]. > My priorities, in order, are: [list five priorities, such as reliability, safety, comfort, low maintenance, accessibility, resale value, or energy efficiency]. > My concerns are: [list your worries]. > > Help me evaluate this decision in four parts: > 1. Create a comparison table using only facts I have provided or facts that can be verified from reliable, current sources. > 2. List the full five-year cost categories I should estimate, including costs people often overlook. Do not invent prices; mark unknown amounts as “needs verification.” > 3. Ask me ten practical questions that would materially change the decision. Ask one question at a time, waiting for my answer before asking the next. > 4. After I answer, give me three scenarios - most financially cautious, balanced, and convenience-focused. Explain the trade-offs in plain English. Do not choose for me. > > Separate confirmed facts, reasonable assumptions, and unanswered questions. If you are uncertain, say so clearly. End with a checklist of what I should verify with an independent expert, official manufacturer information, a mechanic, financial institution, or other appropriate source.

Why this prompt works: It gives the AI a job with boundaries. Instead of asking for a recommendation, it asks for categories, questions, scenarios, and missing information. The instruction to separate facts from assumptions is especially valuable: it makes uncertainty visible rather than allowing a smooth paragraph to disguise a guess. Asking one question at a time also prevents the familiar experience of receiving a long questionnaire and answering none of it. Most important, the prompt includes the option of keeping what you already own. Without that option, AI may quietly assume that buying something new is the goal.

Use Case 2 of 2

Use Case 2: Writing a Professional Decline Email

At 10:12 on Friday morning, David Chen opened an email from a former colleague asking him to join a committee. The request was flattering, and the committee’s work mattered. But David, 64, had already promised himself that he would reduce unpaid obligations this year. He stared at the blank reply box for twenty minutes because every version sounded wrong. “I’m too busy” sounded dismissive. “Maybe later” was not true and might invite another request. A long explanation felt defensive. What David needed was not a clever sentence. He needed a short, respectful email that closed the door without damaging the relationship.

Declining professionally is harder than it appears because people often confuse kindness with excessive explanation. A good decline normally has four parts: appreciation, a clear answer, a brief reason if useful, and a courteous closing. AI can help you adjust the tone, remove accidental harshness, and make the message shorter. It should not manufacture an excuse or make you sound like a corporate announcement.

Here is the exact prompt to use:

Copy and paste this into ChatGPT or Claude:

> Help me write a professional email declining a request. > > The person’s name is: [name]. > Our relationship is: [client, manager, colleague, former colleague, neighbor, professional contact, etc.]. > They asked me to: [describe the request]. > My decision is: [decline completely / decline this time but remain open to a specific future possibility]. > The honest reason, stated briefly, is: [give the real reason in plain language]. > I want the tone to be: warm, direct, respectful, and not overly formal. > I do not want to: sound guilty, invent an excuse, criticize the request, or leave the impression that I am negotiating. > > Write three versions: > 1. A concise version of 60 - 80 words. > 2. A warmer version of 90 - 120 words. > 3. A very direct version of 40 - 60 words. > > Each version must thank the person without excessive praise, clearly decline the request, include my brief honest reason, and end on a courteous note. Do not use phrases such as “unfortunately,” “at this time,” “I regret to inform you,” or “maybe someday” unless they are genuinely appropriate. Do not add commitments I did not make. After the three drafts, explain in one sentence how each version differs in tone.

Why this prompt works: It supplies the relationship, the decision, the honest reason, and the boundaries. Those details matter more than asking AI to “make it sound professional.” The word limits prevent an apology from turning into an essay, while the three versions let you choose rather than accept the first draft. The instruction not to invent an excuse protects your credibility, and the ban on vague phrases helps avoid accidentally inviting continued persuasion.