This week, we’re doing two very practical things with AI: **planning a road trip** and **writing a complaint letter that actually gets results**. Both are situations where you’re juggling details, emotions, and time - and where a calm, tireless assistant can make a noticeable difference in how smoothly things go. These are not “tech person” skills; they’re everyday life skills that happen to use new tools.

Try at least one before Sunday. That is the whole assignment.

Use Case 1 of 2

Use Case 1: Planning a road trip using AI

On a Tuesday evening in late July, around 8:30 pm, Lisa (57) is sitting at her kitchen table in St. Louis with a pad of paper, a glass of iced tea, and her laptop. She and her husband Mark want to drive up to Michigan in September - see some lakes, hit a few small towns, maybe revisit a place they loved in their 20s.

The problem: every time she starts planning, the questions multiply. How many hours of driving per day feels reasonable now that Mark’s back gets stiff if he sits too long? Are they over‑packing the days with stops? What if it rains for two days straight? And which little towns are charming, and which are just tired? She has 12 browser tabs open: maps, hotel sites, random blog posts. She’s exhausted before they’ve even booked a room.

What matters to Lisa is not just “seeing things.” She wants a trip that feels relaxed, with good food, scenic drives, and no “we spent half the day driving in circles because I didn’t realize those two stops were 90 minutes apart.” She wants help thinking it through, not just a list of obvious tourist attractions.

Here is the exact prompt to use:

Copy and paste this into ChatGPT or Claude:

> I want you to act as a road trip planning assistant. > > Ask me your questions first, one at a time, and then build a detailed plan. > > Here is what I know so far: > - Starting city: [your starting point] > - General direction: [e.g., “north toward the Great Lakes”] > - Trip dates / rough timing: [exact dates or “5 - 6 days in September”] > - Travelers: [e.g., “me (57) and my husband (62), no kids, both can walk a mile but don’t want long hikes”] > - Daily driving limit: [e.g., “no more than 5 hours of driving in one day”] > - Budget level: [e.g., “mid‑range: not hostels, not luxury resorts”] > - Interests: [e.g., “scenic drives, small towns, bookstores, good local food, lake views, no nightlife, no big party scenes”] > - Things we don’t want: [e.g., “no chain restaurants, no crowded tourist traps, no steep mountain roads”] > > After you ask your questions and I answer them, create: > 1. A day‑by‑day itinerary with driving time estimates, key stops, and 1 - 2 restaurant suggestions per day. > 2. A list of 3 - 5 “hidden gem” stops that are less touristy but still interesting. > 3. A simple budget breakdown (lodging, gas, food, activities) using reasonable estimates. > 4. A backup plan for bad weather for at least two days (indoor activities or different routes). > 5. A short packing checklist based on the season and destinations. > > Always tell me when something is just an estimate and remind me which details I should double‑check myself (like opening hours, current prices, and road closures).

Why this prompt works:

It gives the AI **structured information** - where, when, who, how long, budget, interests - which is exactly the kind of context it needs to plan well. You’re also telling it what you *don’t* want, which cuts down on generic suggestions. Finally, by asking it to ask you questions first, you force it to clarify your preferences before it starts planning, which usually produces a much better first itinerary.

AI is excellent at brainstorming routes and ideas but **weak on live, factual details** - things like today’s fuel prices, exact hotel rates, and whether a restaurant has closed. Think of the plan as a polished draft, not a final booking sheet. Use it to narrow your choices, then verify all specifics on maps, booking sites, and official attraction pages before you commit. And be careful about over‑stuffed itineraries: if a day looks too packed for your energy level, ask the AI to cut it back and build in downtime.

Use Case 2 of 2

Use Case 2: Writing a complaint letter that actually gets results

Thursday morning, 9:15 am. David (64) is at his desk in his home office outside Atlanta, staring at a half‑written email to his internet provider. In the last three months, his service has gone down repeatedly. He’s spent hours on the phone, had two technician visits, and still deals with outages during the very hours he needs the connection to work.

He’s frustrated and, frankly, insulted by the canned responses and “have you tried unplugging the modem?” scripts. He wants a partial refund and a serious fix, but he doesn’t want to come across as unhinged or be dismissed as “just another angry customer.” His first draft swings between too emotional (“I am fed up and this is unacceptable!”) and too vague (“The service has not met my expectations.”).

Why this matters to David: at this point in life, his time and energy are more valuable than ever. Sitting on hold and writing long emails that go nowhere feels like theft. He wants to be **clear, factual, and firm**, and to send a letter that a real person will take seriously and act on, not just drop into a template reply pile. AI can’t force the company to behave, but it can help him craft something that is much more likely to be read, understood, and escalated.

Here is the exact prompt to use:

Copy and paste this into ChatGPT or Claude:

> I want you to act as a professional customer complaint letter writer. > > I will describe my situation, and I want you to draft a clear, firm, but respectful letter that is likely to be taken seriously and acted on. > > Here is the situation: > - Company: [name of company] > - Product or service: [e.g., “home internet service, 500 Mbps plan”] > - My relationship: [e.g., “customer for 7 years, bills always paid on time”] > - Problem: [brief summary: e.g., “frequent outages over the last 3 months, especially weekday mornings, causing missed meetings and lost work”] > - Specific dates or examples: [list 3 - 5 specific outages, failed appointments, or broken promises] > - Attempts to resolve: [e.g., “three phone calls, two technician visits, one modem replacement, still no stable service”] > - Impact on me: [e.g., “lost income, missed deadlines, personal stress, disruption to household”] > - What I am asking for: [e.g., “a partial refund for the last 3 months, a written plan to fix the issue, and a direct contact person for follow‑up”] > - Tone I want: [e.g., “assertive but polite, factual, no insults, no legal threats unless absolutely necessary”] > > Using this information, write: > 1. A subject line that will make a customer service manager take the email seriously. > 2. A 3 - 5 paragraph letter that: > - States the facts clearly and chronologically. > - Quantifies the impact where possible (time lost, appointments missed). > - Refers to my long‑term relationship as a customer and expectation of fair treatment. > - Clearly states what I want them to do and by when. > 3. A closing that invites a prompt, specific response (not just a generic apology). > > Avoid emotional language that sounds like venting. Focus on facts, dates, and impact, and write in simple, direct sentences.

Why this prompt works:

You’re giving the AI **all the key ingredients** of a persuasive complaint: a clear problem, concrete examples, documented attempts to resolve, and a specific remedy you’re seeking. That allows it to structure the letter in a way that a human reader can follow and take action on. You’re also telling it what tone to use and what to avoid, which keeps it from sounding either too aggressive or too weak.

AI will happily add confident‑sounding phrases that **might not be accurate** - for example, “as mandated by consumer protection law” or “this violates your terms of service” - even if you never mentioned those. Do not let it invent legal claims or quote policies you haven’t verified yourself. Read the draft slowly and remove anything that isn’t strictly true or that you can’t back up. And remember that even the best letter can’t guarantee results; it improves your odds, but you may still need to follow up by phone or escalate to a supervisor or regulator if the company stonewalls.