AI Planned My Family’s NYC Adventure — Here’s How It Went

Planning my family trip to New York City was exciting but overwhelming, with many things to do and ways to get there. So when my family flew through NYC on our way to Connecticut, I turned to ChatGPT to bring structure to our loose plans.

With short windows on our arrival and departure days, we really had to make the most of our time to sightsee. Using artificial intelligence helped me create a personalized and flexible itinerary that was easy to follow and offered helpful insights that would have been more difficult to figure out independently.

Read on to see how I planned our days in NYC using AI, the expectations versus reality, and what ChatGPT got right and wrong.

Planning Our Trip With ChatGPT

I had an idea of what I wanted to do with our short time in New York, but I needed help organizing all the details, including how to structure our days, arrange transportation, and compare costs.

In my experience, ChatGPT works best when you break a larger query down into tasks and then summarize it once you’ve established the details. My main query was to plan how we’d get from LaGuardia Airport (LGA) to Southbury, Connecticut, and back, with tourist stops along the way. I separated transportation and touring, starting with transportation and then adding sightseeing after we nailed down how we’d get from LGA to Connecticut and back with a stop in Manhattan.

Transportation

I gave ChatGPT some initial transportation plans to work through:

• Arrive at LGA around 1 p.m.
• Visit Manhattan on our way to Connecticut
• Get to Southbury, Connecticut, with a rental car by 8 p.m. on our arrival day
• Spend more time in NYC on our departure day with an evening flight

I told ChatGPT I needed the most help with transportation from Manhattan to Southbury, picking up a rental car somewhere along the way, and preferring to avoid driving in New York.

ChatGPT gave me a timeline and options for getting around (including a few tourist spot suggestions), then gave me a recommendation for an overall plan:

Step 1: Get from LGA to Manhattan (1:00 p.m. to 1:30 p.m.)

Since you have luggage and kids, a taxi, ride-sharing app, or shuttle might be more convenient than public transit.

• Taxi/Ride-Sharing App (Uber/Lyft)
  • Cost: $40-$60 (plus tolls, traffic-dependent).
  • Travel time: 25 to 40 minutes to Midtown.
  • Easiest option, especially with bags.
• Express Bus + Subway
  • Take the LaGuardia Link Q70-SBS (free) to Jackson Heights-Roosevelt Av Station.
  • Transfer to an E, M, F, R, or 7 train to Midtown.
  • Cost: $2.90 per person (kids under 44” ride free).
  • Travel time: 45 minutes.

Step 2: Explore Manhattan (1:30 p.m. to 5:00 p.m.)

With about 3.5 hours left, pick a manageable plan based on your family’s interests. Some ideas:

• Central Park and Times Square: Walk, see the famous spots, grab a bite.
• Empire State Building or Top of the Rock: Quick iconic view.
• Museum Visit: The Natural History Museum or MoMA are good for kids.

Step 3: Pick Up a Rental Car Before Heading to Southbury

Since you need a car for Southbury, picking it up in Manhattan or an outer borough is best.

Best Rental Car Pickup Options

1. Manhattan (Midtown or Upper West Side)
   • Pros: Easy access before leaving the city.
   • Cons: Expensive ($120+/day), heavy traffic.
   • Best For: Convenience if you don’t mind driving out of the city.
2. Near Grand Central Station (East Side)
   • Best Locations: Hertz, Enterprise, Avis (close to FDR Drive for easy exit).
   • Take the Metro-North train to Stamford or Bridgeport, then drive to Southbury.
3. Newark or White Plains Airport Rentals
   • Newark: Take a NJ Transit train from Penn Station to EWR, then rent.
   • White Plains (HPN): Take a Metro-North train to White Plains and rent there.
   • Pros: Avoid NYC traffic before driving.
   • Cons: Requires a train ride first.
4. Stamford, CT (Best for Easiest Drive)
   • Take a Metro-North train from Grand Central to Stamford (35-45 min).
   • Rent a car at Stamford Station (Avis/Hertz).
   • Pros: Avoid city driving, cheaper rentals.
   • Cons: Need to carry luggage on train.

Step 4: Drive from Manhattan or Stamford to Southbury

• From Manhattan: 2 hours (depending on traffic).
• From Stamford: 1 hour.

Best Overall Plan

1. Take a taxi/ridesharing app from LGA to Manhattan (1:30 p.m.).
2. Spend 3.5 hours in Manhattan.
3. Take Metro-North from Grand Central to Stamford (5:00 p.m.).
4. Pick up a rental car in Stamford (6:00 p.m.).
5. Drive to Southbury (1 hour).

This plan minimizes NYC driving and saves rental costs. Let me know if you need more options!

The most helpful aspect of our ChatGPT transportation plan was the guidance on rental cars. I had no clue where to get a rental car, and didn’t want to drive anywhere near NYC during rush hour when we planned to leave. It was great to know the pros and cons, including cost estimates.

We chose the Stamford, Connecticut, rental car option because we could take the Metro-North train from Grand Central to Stamford, then rent a car at Stamford Station. I felt this was the best option because when I looked into the details, I saw the rental office was at the train station, so we wouldn’t have to use a ride-sharing app to get from the train to pick up our rental car.

Activities

With our transportation plan in place, I then asked ChatGPT to map out how we’d spend our time in Manhattan on our arrival and departure days.

I like using ChatGPT to form a more concrete plan for random ideas, and that’s what I did here. I confirmed our transportation plans and gave it a list of sights we’d like to see. I knew we couldn’t get to it all, but I wanted ChatGPT to figure out what activities would fit well together in terms of the time we had available and grouping activities by location:

I did some back and forth with ChatGPT to clarify our preferences, including our Stamford arrival time, ensuring we’d get there before the rental office closed at 7 p.m. I also asked it to include alternative plans in case we found ourselves ahead or behind schedule.

Here’s the plan we landed on:

Note: This plan contains some inaccuracies, including pricing and timing.

Arrival Day: LGA → Manhattan → Stamford → Southbury

1:00 p.m. – Land at LGA and Head to Grand Central (1:45 p.m.)

Uber/Taxi: $40-$60 (25-40 min)
Drop-off: New York Public Library (5th Ave and 42nd St)

---

1:45 p.m. – 2:30 p.m.: New York Public Library

• Explore:
  • Rose Main Reading Room
  • Patience and Fortitude statues
  • Quick exhibits visit

Walk to Rockefeller Center (2:30 p.m. – 2:45 p.m., 15 min walk)

---

2:45 p.m. – 3:45 p.m.: Rockefeller Center and Nintendo Store

• Rockefeller Plaza and Ice Rink (quick photos)
• Nintendo NYC Store (30 min, extended from 15 min)

Walk to Grand Central (3:45 p.m. – 4:00 p.m., 15 min walk)

---

4:00 p.m. – 4:30 p.m.: Early Dinner at Grand Central

• Quick options:
  • Shake Shack ($10-$15 per person)
  • Zaro’s Family Bakery ($5-$10 per person)

4:45 p.m. – 5:30 p.m.: Train to Stamford (Adjusted for 6:15 p.m. Arrival)

• Metro-North Fare: (35-45 min train ride)

6:15 p.m. – 7:15 p.m.: Pick Up Rental Car and Drive to Southbury

---

Departure Day: Southbury → Stamford → Manhattan → LGA

8:00 a.m. – 9:30 a.m.: Drive to Stamford and Drop Off Rental Car
9:30 a.m. – 10:30 a.m.: Train to Grand Central ($12 per adult, $6 per kid)

---

10:30 a.m. – 11:00 a.m.: Central Park Carriage Ride (30 min)

59th St and 6th Ave (Carriage Ride Start Point)
Cost: $120 total for 5 people

• Route:
  • Pass by Wollman Rink, Bethesda Terrace, and The Mall
  • Scenic loop around Central Park

Walk to Grand Central (11:00 a.m. – 11:15 a.m., 15 min walk)
Backup: Uber (5 min, $15).

---

11:15 a.m. – 11:45 a.m.: Train to Strand Books

Take 6 train from Grand Central to Astor Place ($2.90 per person), 5 min ride + 5 min walk.

11:45 a.m. – 12:30 p.m.: Explore Strand Books

828 Broadway and 12th St

• Browse rare books, souvenirs, and gifts
• Visit the Rare Book Room (third floor)

Walk to Prince Street Pizza (12:30 p.m. – 12:50 p.m., 0.5 miles, 10 min walk)

---

12:50 p.m. – 1:10 p.m.: Stop at Prince Street Pizza (Quick Bite)

27 Prince St

• Must-try: Spicy Pepperoni Square Slice (about $6 per slice)

Walk to Hamburger America (1:10 p.m. – 1:30 p.m., 0.8 miles, 20 min walk)

---

1:30 p.m. – 2:15 p.m.: Lunch at Hamburger America

Hamburger America (155 W Houston St)

• Hamburger meal ($10-$20 per person)

Optional: Decision After Lunch

• Option 1: Walk to Ghostbusters Firehouse (1.7 miles, 16 min) 14 N Moore St, Tribeca
• Quick 10-minute stop for photos
• Then continue to Staten Island via subway or ride-sharing app

Option 2 (Shortcut): Go straight from Hamburger America to Staten Island (2 miles, 40 min walk or subway).

Walk from Ghostbusters Firehouse (or Hamburger America) to Staten Island Ferry Terminal (2:30 p.m. – 3:15 p.m.)

---

3:30 p.m. – 4:30 p.m.: Staten Island Ferry (FREE, Round-Trip)

Views of Statue of Liberty and Manhattan Skyline

Return to Battery Park (4:30 p.m.).

---

5:00 p.m. – 6:00 p.m.: Ride-Sharing App to LGA (Default Option)

Uber Cost: $50-$70 (30-40 min)

---

Backup Plan: If Running Late, Take an Uber from Staten Island to LGA

Estimated Time: 50-60 min
Estimated Cost: $70-$90
Decision Point: If short on time, take Uber from Staten Island instead of returning on ferry.

---

8:12 p.m.: Flight Departs

AI Travel Plans vs. Reality: What Actually Happened

Our flight was on time, and we left LGA at 1 p.m. While on the plane, I checked our real-time transportation options to the New York Public Library using Google Maps. I found that ride-sharing app options were closer to $75 (more than the $40 to $60 ChatGPT estimated) and would save us less than half an hour compared to public transportation.

ChatGPT had recommended we use a ride-sharing app, but offered the LaGuardia Link Q70-SBS and subway as an alternative, and Google Maps agreed. To get from LGA to Manhattan, we took the bus and the F train on the subway, which dropped us off near the subway and cost less than $15.

Once we got off the subway, I followed Google Maps and subway signs to reach the New York Public Library. This was a high-priority stop, as I wanted to take the kids to see the rare book treasures in the Polonsky Exhibition.

Our AI-generated schedule helped us stay on time. We left the library just 10 minutes behind schedule, walking to Rockefeller Center for a quick photo stop before visiting Nintendo New York. ChatGPT estimated we’d walk there in 15 minutes, but it took about 25 because we stopped for a hot dog on the way.

Next was Nintendo New York, which is in Rockefeller Center.

I budgeted an hour for this visit with ChatGPT because I knew the kids would play with the interactive features and consoles as long as we’d let them!

Since we’d grabbed hot dogs on the way to Rockefeller Center, no one was hungry for the early dinner at Grand Central ChatGPT suggested. Instead, we did an express visit to FAO Schwarz — including a few minutes of piano dancing — since it was just steps away from Nintendo New York and we had some time before getting on the Metro-North train.

We wrapped up at FAO Schwarz at 4:30 p.m., just 15 minutes before we’d planned to board a Metro-North train to Stamford from Grand Central Terminal. It took us about 25 minutes to walk to Grand Central — a little slower than the 15 minutes ChatGPT had estimated.

We were a little behind schedule, but ChatGPT helped me build a buffer of about 45 minutes for picking up our rental car, so we still had time. We hopped on a Metro-North train that departed just 5 minutes after we boarded. Since we cut it close and rode the train at rush hour, few seats were left, but it was only about a 45-minute ride.

We arrived at the Stamford Station Avis rental office at 6:20 p.m., almost perfectly on time for our planned 6:15 p.m. arrival. That was critical because the rental office closed at 7 p.m., and I didn’t want to drive in an unfamiliar place late at night anyway.

We stopped for dinner near Wyndham Southbury, our destination, and then spent a few days visiting family there before returning to New York City.

We had a little more time on our departure day, but got a late start. Our original plan with ChatGPT was to leave at 8 a.m. and arrive at Grand Central and Central Park by 10:30 a.m. ChatGPT didn’t include any time in our plans for getting from Grand Central to Central Park — about a 20-minute subway ride or a 30-minute walk. But that detail didn’t matter in the end, as 8 a.m. proved too ambitious a departure time for us, and we arrived at Grand Central at 12:30 p.m. rather than 10:30 a.m.

While riding the Metro-North train and knowing we’d have to cut something out of our plans, I used ChatGPT to pivot our plans on the fly. Though we wanted to do a Central Park carriage ride, it was far from our other planned activities, and we decided it would be a better experience on a future trip with fewer time constraints. I’d promised my kids they could pick out books at Strand Bookstore, and my husband was eager to check off his food stop list, so those plans were not negotiable.

We stayed within Grand Central to grab a snack and then hopped on the subway to Strand Books to pick out plane reads.

Then, we walked to Prince Street Pizza, which took us about 20 minutes, a little longer than the 10-minute walk ChatGPT planned. But ChatGPT budgeted a 20-minute stop here, and we didn’t need that long to grab a slice and go. This was just an appetizer stop before heading to Hamburger America!

From there, we walked to Hamburger America, which was on my husband’s must-stop list as a fan of founder George Motz. We had a brief sit-down lunch outside, then checked our time against the ChatGPT plans.

We were just about on time for our schedule, so we hit the “Ghostbusters” firehouse nearby — about a 15-minute walk. ChatGPT budgeted a 10-minute stop here, but we didn’t stay that long — just enough for a few quick photos.

After the “Ghostbusters” firehouse, we had time to ride the Staten Island Ferry. ChatGPT had plans for walking 45 minutes and getting on the 3:30 p.m. ferry, but we rode the subway instead, which saved us time and complaints from the kids. We arrived in time for the 4 p.m. ferry.

We got on the starboard side of the boat near a window to see the Statue of Liberty.

Riding the ferry gave us a good view of Lower Manhattan from the water.

We rode the ferry because it passes by the Statue of Liberty and Ellis Island. It was a great setting for us to discuss the Statue of Liberty, the history of immigration in the U.S., and some of our family’s German ancestors who came through Ellis Island.

We rode the ferry round-trip and returned to Whitehall Ferry Terminal at 5 p.m., right on time. I’d planned with ChatGPT to use a ride-sharing app to get from there to LGA, knowing we’d probably be ready to relax on our ride to the airport rather than navigate public transportation at rush hour.

When I pulled up ride-sharing app options, they were around $100 for an XL vehicle to get all 5 of us to the airport. That was much higher than ChatGPT’s estimated ride cost of $50 to $70. Rush hour surge pricing was in effect, which ChatGPT hadn’t accounted for.

We were all a little worn out, and following a Google Maps public transportation plan would have added 30 to 45 minutes to our journey, so we splurged on an Uber.

It was worth it, because I was eager to make it to the Chase Sapphire Lounge at LGA, free for me and up to 2 guests with my Chase Sapphire Reserve^®. Also, I had an up to $15 monthly Uber credit from The Platinum Card^® from American Express to knock off some of the cost of our ride. (To receive this credit, link your Amex Platinum card as a payment method on your Uber or Uber Eats account in the Wallet section of the Uber app and the funds will be deposited into your account automatically on the first of each month.)

Bottom Line:

We didn’t strictly follow our ChatGPT tour plan, but it helped me map out our transportation and must-do activities with a realistic schedule and logical route, including walking and transit with built-in buffer times.

What AI Got Right (And Wrong)

ChatGPT did a great job making a realistic plan for a NYC visit with a complicated itinerary. I threw some basic facts at it, like where we needed to be when, and a list of what we wanted to do, and it came back with helpful suggestions. It made what would have been complicated into a simple plan. For example, I would have spent a lot of time finding an easy-access rental car location on our way from New York to Connecticut, but that took just seconds with ChatGPT.

It was also helpful for changing plans on the fly, when it helped me realize that a Central Park carriage ride was too far out of our way with limited time, and we’d be better off skipping it so we could continue our other plans much closer together.

Although ChatGPT struggled with predicting exactly how long it would take us to walk or transit to a location, it overestimated how long we’d spend at most spots and built in buffers. That made it so we could stay on track with the schedule it created for us.

ChatGPT was almost always wrong with exact pricing, especially for ride-sharing apps, but it did help me get an idea of ballpark costs that I later confirmed. It estimated we’d spend $146 to $166 on train and ride-sharing, but we actually spent $204.

I found it extremely useful to have ChatGPT create a schedule, notes, and time and cost estimates so I didn’t have to research and type it myself. I just had to confirm information and follow the plan!

Tips for Planning Travel With AI

We learned a great deal about ChatGPT’s capabilities for travel planning, so here are some expert tips for the next time you consult AI.

1. Break Up Big Questions

Rather than a big, vague command like “plan my NYC trip,” I asked for specific answers about transportation and organizing our sightseeing priorities.

2. Use AI to Logically Group Activities

I asked ChatGPT to group our sightseeing by location to minimize backtracking, which helped us save time and effort. I also asked it to include transportation details so we’d have an idea of how we’d get around, even if we didn’t follow its recommendations exactly.

3. Be Specific About Priorities

I told ChatGPT which of our activities were not negotiable, so it built the rest of our days around those.

4. Fact-Check Everything

AI is a great starting point, but it’s not good with factual details. You need to double-check opening hours, rental car availability, time estimates, train schedules, directions, and current ride-sharing app prices.

5. Use AI To Identify Alternatives

One of the most helpful aspects of using ChatGPT for our NYC tour was the suggestion to rent a car in Stamford instead of Manhattan to avoid traffic and save money.

6. Tell AI Who You’re Traveling With

Including this information can help AI make recommendations based on the number of travelers, ages, mobility needs, and general travel style. I told ChatGPT we were traveling as a family of 5 with kids, which helped us avoid overly ambitious activities and too much walking.

7. Ask for Backup Plans

It’s always good to be flexible, and AI can help you build in buffer times and alternatives if you’re behind schedule or find yourself with extra time. Since it was in our plans and we had the time for it, we got to squeeze in a side quest at the “Ghostbusters” firehouse.

Final Thoughts

Would I use AI to help me plan travel again? Absolutely. While it didn’t create a perfect plan, it made sense of all the ideas and needs I threw at it to assemble a detailed schedule that was easy to follow.

I’ve used ChatGPT to help me plan road trips with scenic stops, weigh pros and cons of various travel options, and create touring schedules like this one we used in NYC, and I’ll continue to use it as a resource for organizing our travel plans.