One of the hottest topics of the moment is the conversation around artificial intelligence (AI.) If you’ve been paying attention to this conversation, you likely know that Artificial Intelligence programs have made enormous strides forward in just the past few years. These advances have caused disruption and innovation in equal measure in many fields. These advances prompted discussion in the Model Railroader offices, our primary question being, naturally, what can these programs do for us, and for our hobby?
Curious as to the capabilities of these programs as they relate to model railroading, I started to experiment. Using a program known as ChatGPT, a web-based AI chatbot, I began to test the knowledge and capacity of this program. By asking it questions, ChatGPT generates responses based on the specified parameters of the question – more open-ended questions yield more open-ended responses, and more specifically-defined questions result in more specific answers. The program uses a generative learning model, which pulls from a broad spectrum of resources across the web to generate responses without human input.
Getting Started
To start, I asked an open-ended question to see in what direction the program would take me. My first query: “Generate me a model railroad trackplan layout.”
The response I received:
Let’s start with the basics: a 4×8 layout with a rural countryside theme. Okay, maybe no points for originality, but it’s certainly a workable suggestion. From there, the program specifies what goes where on the layout. In the bottom left corner, we have a small village with a train station. Our mainline curves around the outside of the village and crosses a bridge over a river before entering a tunnel. I can easily picture what is being described in my mind, so it’s plausible.
From there, the track splits, with one path diverging into a mountainous region with a mining operation while the other continues to an urban area with a passenger station and a switching yard.
Well, okay. Seems a bit dense for a 4×8 layout to me, but then again, scale was never specified, which makes a difference. Without that having been specified, how was I to know what scale to use on this layout? So, I asked:
A 4×8 HO scale layout with a rural countryside theme. Again, perhaps no points for originality, but some of the details are well-rendered, and I have no problem picturing this layout in my mind. From there, I posed more questions about the layout, including era, benchwork, and what motive power would be appropriate for a layout matching this description.
The suggested locomotives are certainly viable recommendations. There are of course plenty of options available in the model railroading market for late-period steam locomotives, SD40s, and other such locomotives.
Narrowing results
Wanting to see what would change when I specified the scale in my initial inquiry, I entered almost exactly the same question, but this time including my desired scale (HO.) The result I got back:
As you can tell, the result is almost exactly identical to the first question I asked, and the response is structured in the same manner as that which preceded it.
Specifying response structure
Wanting to change things up and further test the capacity of this program, I changed how I phrased my question in a new query: “Generate me an N scale model railroad layout with detailed description including bullet points.”
The response I received:
I found myself much more impressed by this response. By structuring the parameters of the response I wanted, I seem to have more clearly defined for this program what individual details were needed without actually specifying them. The bullet point structure of this response gives clearly defined individual features, including track, lighting and operations.
Finally, to round out this inquiry, I asked “What motive power would be appropriate on this layout, considering the specifications you have provided me?”
The response was as follows:
Other articles about AI programs have pointed out their tendency to uncritically share incorrect information. I shouldn’t be surprised that the program didn’t know that an NW2 isn’t a steam locomotive.
Generating an image
I also wanted to test out the capacity of similar AI programs focused on image generation. I fed the details into a program called Fotor, and from the prompt “Model railroad layout trackplan, birds eye view, HO scale, mountainous, with tunnel, mining operation and city” I received these images:
Now, if you don’t look closely, these are interesting enough images, but hardly trackplans in the style any model railroader would recognize. Upon closer inspection, though, one might notice a locomotive and a building morphing into one entity, or nonsensical track architecture, or bridges leading into the sheer sides of cliffs, no tunnel entrance in sight. Hardly a track plan, nor a logical interpretation of a model railroad layout. From looking at these images, it would seem the program seems to know what to replicate, but not how to do so.
Considering these results, I found myself underwhelmed. That’s not to say that these are useless programs; there are applications. For example, a beginner modeler could use ChatGPT as a tool to generate inspiration and useful information for a prospective layout. However, it’s ultimately just an aggregator. This program, and others like it, pull information from thousands of different places, finds what occurs most frequently, and regurgitates that to the user without any process to verify the information. Essentially, a user is hav a conversation with a search engine. As we’ve seen, there are ways to better define and structure the response you receive, but ultimately, the information is coming from the same place. Again, this is not to say that the program is useless. But perhaps the truly revolutionary program some believe this to be is as of yet unrealized. For now, it’s a lowest common denominator machine.
Interested in learning more about potential AI use cases for model railroaders? Check out the links below! – Ed.
ChatGPT use cases for the model railroader
After a year of change in the field of LLMs, a look at the progress made and what these new developments mean for model railroaders.
Revisiting AI for model railroaders
A comparison between ChatGPT and Google’s Gemini text-based LLM models, and an examination of the image generation results from Microsoft’s Bing image generator.
Designing structure interiors with artificial intelligence
An exploration of the practical application of image generation AI programs, and a practical application of AI generated structure interiors on Model Railroader‘s Fremont Mills layout.
Proves you can have all the information you need right in front of you and still get the wrong answer. AI is not human intelligence. It try’s to mimic our thinking. That “spark” we have in our brains is hard to produce in a computer no matter how much information is available.
One item caught my eye – the specification of a DC loco. Wonder how that popped up. Also, the description of the initial 4 x 8 layouts reminded me of a similar description in an April issue of MR where in that size the track ran from NYC to the West Coast across rivers, plains, bridges and mountains, through industrial and rural towns, etc. A bit crowded.
“I shouldn’t be surprised that the program didn’t know that an NW2 isn’t a steam locomotive.” Actually, I’d say you should be surprised–there is plenty of information out there on the web that would make it clear the NW2 is a diesel. But that is the problem with ChatGPT (at least so far)–it makes grievous mistakes. I’ve seen it be given a well-written set of rules to a game, be asked a rules question, and come up with an answer that was clearly, absolutely wrong; an unquestionable violation of the rules. The game wasn’t Chess, but if it was, it would be like ChatGPT stating that a Bishop could move like a Knight. Nothing, not even a Queen, can move like a Knight. The rules to Chess aren’t hard–unless you are ChatGPT. Definitely not ready for prime time.
It works good generating code for different DCC systems.
This is such an interesting concept, using ChatGPT to inspire a layout design. Plain language tools like ChatGPT are more about summarization, and reconfiguration of known data. It’s not very good at pulling up technical facts. I’ve heard it described this way, it’s an English professor, not a librarian.
I use it to inspire a different perspective on a topic, find a new line of questioning, or as a brainstorming tool. Knowing full well that factually it may still be suspect, at best. When using ChatGPT you still need to think in the abstract to get the most out of it.
Ben, I like the summary you provided: An English professor, not a librarian. That’s very good.
(I went down the rabbit hole for more detail)
“Let’s explore the whimsical fairy garden model railroad. Railroads have purpose. They exist to efficiently move goods and passengers from place to place. Create a bulleted list of 10 ‘industries’ (fairy industries with one or two ‘population centers’ that would be a source of passengers) that you would find on this fairy-themed model railroad. Include the type of goods that the fairy industry produces and what kind of railroad car that goods would use to be moved by rail.”
– Dragon’s Forge – produces enchanted metal goods and weaponry that would be transported in flatcars or gondola cars.
– Fairy Dust Distillery – produces enchanted potions, elixirs, and other magical substances that would be transported in tank cars.
– Unicorn Ranch – produces enchanted milk, cheese, and meat that would be transported in refrigerated boxcars, as well as moving the unicorns in livestock cars.
– Rainbow Flower Farm – produces enchanted flowers with rainbow hues that would be transported in covered gondola cars.
– Starlight Glass Factory – produces enchanted glass with shimmering colors that would be transported in boxcars.
– Pixie Lightbulb Manufacturing – produces enchanted light bulbs that would be transported in boxcars.
– Magic Inkwell Company – produces enchanted ink and quills that would be transported in boxcars.
– Mystic Gem Mine – produces enchanted gemstones and crystals that would be transported in hopper cars.
– Mermaid Pearls and Coral Company – produces enchanted pearls, coral, and other undersea treasures that would be transported in refrigerated boxcars.
– Unicorn Hair Studio – produces enchanted hair for wigs, beards, and other costume accessories that would be transported in boxcars or flatcars.
As for population centers, here are a couple of ideas:
– The Enchanted Forest Village – home to the forest fairies, who would provide a steady stream of passengers for the railroad.
– The Fairy Tale Castle – home to the fairy tale characters like Cinderella, Snow White, and the Three Little Pigs, who would also use the railroad for transportation.
Well, if you ask basic questions, you’re going to get basic answers. The answers you got about the HO scale layout are…exactly the advice that Model Railroader gave me when I was 13 years old! *grins*
To get good results, you have to be specific, and…to demonstrate that ChatGPT is actually being *creative*, you can mix it up a bit!
************ QUESTION ************
“I’m a moderately experienced model railroader and I’ve just moved into a new house and I don’t have a lot of room for my new layout. I have a space 3 feet wide, and 11 feet long on one wall, and 3 feet wide and 9 feet long on the far end of the room, along the perpendicular wall. There’s a window in the middle of the first wall, so…no mountains along the ‘long’ side because it would block the window, and be a fire code violation.”
“This is my 4th model railroad, so I’ve done the typical ‘mid-century’ and ‘1940s’ layouts with steam and diesel locomotives mixed in.”
“I’m looking for ideas on something new and creative that will use N-Scale (1/160) to maximize what I can put on my new model railroad.”
“Give me some suggestions on what theme I could use with this model railroad that is unusual, and photogenic, because I want to submit lots of photos and articles to Model Railroader Magazine during the build and after it’s complete.”
“To the best of your ability, answer in the comedic delivery style of 60 Minutes correspondent, Andy Rooney.”
*********** ANSWER ************
“Well, well, well, it looks like we’ve got ourselves a bit of a conundrum here, folks. Our model railroader friend is in a bit of a tight spot, ain’t he? 3 feet wide and 11 feet long on one wall, and 3 feet wide and 9 feet long on the other. Sounds like a perfect spot for a miniature train track if you ask me. But here’s the catch, there’s a window in the middle of the first wall, and we can’t go around blocking windows now, can we? That would be a fire code violation! I mean, who needs safety when you can have trains, am I right?”
“Now, our friend here says he’s done the typical “mid-century” and “1940s” layouts with steam and diesel locomotives mixed in. I don’t know about you folks, but I’m getting a little bored just thinking about it. I mean, how many times can we recreate the same old locomotives and railway lines? It’s time for something new, something unusual, something that will make our friend’s model railroad stand out from the rest.”
“So, here’s what I’m thinking. How about a model railroad based on a futuristic, sci-fi theme? You heard me right. Picture this: a sleek, high-speed bullet train zipping through a city of the future, complete with towering skyscrapers, monorails, and flying cars. It’s like something straight out of a sci-fi movie! And let’s not forget about the robots and aliens, because what’s a futuristic city without those?”
“Or how about a model railroad based on a fantasy world? Picture this: a magical kingdom complete with wizards, knights, and dragons, all traveling by train. You could have a castle, a dungeon, and even a wizard’s tower, all connected by a magical railway line. It’s like Hogwarts Express, but on steroids!”
“And finally, how about a model railroad based on a real-life location? Let’s say our friend here is a big fan of New York City. Why not recreate the city in miniature, complete with skyscrapers, Central Park, and even a subway line? It’s like a love letter to the Big Apple, but on a much smaller scale.”
“So, there you have it, folks. Some creative ideas for our friend’s model railroad. Whether he chooses a futuristic, fantasy, or real-life theme, one thing’s for sure – his model railroad is going to be the talk of the town. Now if you’ll excuse me, I’m off to play with my own model railroad. All aboard!”
********** ALTERED THE LAST LINE *********
“Answer in the style of Robin Williams playing Mork from Ork”
******** ANSWER ********
“Nanu nanu my fellow model railroader! I sense a challenge here, and I’m ready to take it on like an alien warrior.”
“So, you’ve got a narrow space to work with, but fear not! With N-Scale, you can fit a whole universe in that tiny space. Let’s start by thinking outside the boxcar. How about a post-apocalyptic world where trains are the only means of transportation left? You could build a gritty wasteland landscape with abandoned buildings, rusty tracks, and derailed cars. Add some weathering effects and you’ll have a photogenic scene straight out of a Mad Max movie.”
“Or, if you want to go for something more whimsical, how about a model railroad that’s also a fairy garden? You could build a miniature forest with tiny houses and magical creatures, and have a train running through it. Add some LED lights and you’ll have a magical scene that will make everyone smile.”
“But wait, there’s more! How about a model railroad that’s also a time machine? You could build a diorama that spans different eras, from the Wild West to the Space Age. Add some historical accuracy and you’ll have a photogenic scene that will make history buffs swoon.”
“Whatever theme you choose, remember to have fun with it. That’s what model railroading is all about, after all. And if you need any more ideas, just say the word and I’ll be back faster than a speeding bullet train. Nanu nanu!”
(I went down the rabbit hole for more detail)
“Let’s explore the whimsical fairy garden model railroad. Railroads have purpose. They exist to efficiently move goods and passengers from place to place. Create a bulleted list of 10 ‘industries’ (fairy industries with one or two ‘population centers’ that would be a source of passengers) that you would find on this fairy-themed model railroad. Include the type of goods that the fairy industry produces and what kind of railroad car that goods would use to be moved by rail.”
– Dragon’s Forge – produces enchanted metal goods and weaponry that would be transported in flatcars or gondola cars.
– Fairy Dust Distillery – produces enchanted potions, elixirs, and other magical substances that would be transported in tank cars.
– Unicorn Ranch – produces enchanted milk, cheese, and meat that would be transported in refrigerated boxcars, as well as moving the unicorns in livestock cars.
– Rainbow Flower Farm – produces enchanted flowers with rainbow hues that would be transported in covered gondola cars.
– Starlight Glass Factory – produces enchanted glass with shimmering colors that would be transported in boxcars.
– Pixie Lightbulb Manufacturing – produces enchanted light bulbs that would be transported in boxcars.
– Magic Inkwell Company – produces enchanted ink and quills that would be transported in boxcars.
– Mystic Gem Mine – produces enchanted gemstones and crystals that would be transported in hopper cars.
– Mermaid Pearls and Coral Company – produces enchanted pearls, coral, and other undersea treasures that would be transported in refrigerated boxcars.
– Unicorn Hair Studio – produces enchanted hair for wigs, beards, and other costume accessories that would be transported in boxcars or flatcars.
As for population centers, here are a couple of ideas:
– The Enchanted Forest Village – home to the forest fairies, who would provide a steady stream of passengers for the railroad.
– The Fairy Tale Castle – home to the fairy tale characters like Cinderella, Snow White, and the Three Little Pigs, who would also use the railroad for transportation.
This is very interesting, and definitely something I’d like to look further into. The practical application here seems obvious once you consider an individual modeler may be interested in pursuing a specific interest or theme, such as the one you used.