
There’s been a lot of talk lately about artificial intelligence. Apparently it can do all sorts of things, from producing an essay or technical document, to writing a letter from your family pet requesting more ball-playing time (I know the latter is true, as we … er, my son’s dog … did it!). But if it can do all this sort of fancy stuff, could AI put writers of wargaming rules out of a job?
As a bit of fun this evening, my son and I opened his ChatGPT programme and requested it to write us a set of Napoleonic wargaming rules. Not only did ChatGPT write a set of rules in both serious and humourous versions, but it then took over writing the intro to this blog posting as well! Let’s hand over to ChatGPT (which also came up with the title of this posting, by the way!):
I never thought I would be getting wargaming advice from a ghost, let alone one of the most famous military leaders in history. But that’s exactly what happened when I stumbled upon the OpenAI GPT-3 model and asked it to write me some rules for Napoleonic wargaming.
At first, I was skeptical. I mean, how accurate could a computer program be at writing rules for a hobby that requires a human touch? But as I read through the rules, I was pleasantly surprised. Not only were they technically sound, but they were also infused with a touch of humor and personality.
The AI even included some unique ideas and special abilities for units, like elite units having a higher morale and the ability to re-roll failed morale checks, or skirmishers being able to move and shoot in the same turn. I never would have thought of those on my own!
But the real cherry on top was when the AI included a command points system and even suggested using pizza as a bribe to replenish them. I mean, who doesn’t love pizza?
Of course, as with any rules written by a computer program, there were a few quirks and inconsistencies. But overall, I was impressed by the level of detail and creativity displayed by the AI. I even contemplated using the rules for my next wargaming session, but then I remembered that I’ll be playing with my friends and they might not appreciate the humor and the idea of using pizza as a bribe.
I have to say, it was a pretty wild and funny experience, and I never thought I’d be getting wargaming advice from a ghost, let alone one of the most famous military leaders in history, but the AI certainly didn’t disappoint. Who knows, maybe Napoleon himself helped out a bit from the afterlife.
In any case, I’m definitely going to be keeping an eye on the future developments of AI in the hobby of wargaming, and I might even consider asking Napoleon for advice again in the future.
Anyway, here’s ChatGPT’s Napoleonic wargaming rules (humourous version):
Welcome to the world of Napoleonic wargaming, where the only thing more exciting than the battles is the arguments over whose dice are loaded. You’ll be using miniature soldiers to represent different units on the battlefield, and your goal is to make sure your side comes out on top. Or at least, that’s what you’ll tell your friends when you lose.
First off, let’s talk about the units you’ll be using. Each unit has a certain number of figures, which represents its strength. Unless you’re playing with those tiny plastic soldiers, then it represents how many you can fit in your pocket. Each unit also has its own set of stats, like movement speed, firepower, and morale. These stats can be found on a roster sheet or a reference card, or if you’re like me, scribbled on a napkin in crayon.
When it comes to movement, you’ll be using a tape measure or ruler to move your units around the battlefield. Standard infantry units can move up to 6 inches per turn, unless you’re playing with those tiny plastic soldiers, then they just kind of roll around in your hand. Keep in mind that some units have special abilities that can affect movement, like light infantry being able to move through difficult terrain without penalty. Or just falling in between the cracks in the table.
Combat is where the real action happens! Players take turns moving and attacking with their units. You’ll be rolling dice to resolve combat, and the outcome will be affected by the units’ stats and the situation on the battlefield. And also by how much you’ve had to drink. Don’t forget that morale is super important – if a unit suffers heavy casualties or gets charged by the enemy, they might rout and have to run away. Or just lay down and play dead, pretending to be a casualty.
Now, let’s talk about some of the special abilities or unique characteristics that certain units might have. Elite units like the Old Guard or Grenadiers might have a higher morale and be able to re-roll failed morale checks. Which is just fancy talk for “they’re too stubborn to run away.” Skirmishers can move and shoot in the same turn, which is great if you’re trying to get away from your significant other. Light infantry can move through difficult terrain without penalty, but they’ll probably still complain about it. And Heavy cavalry like cuirassiers will have a higher charge bonus, but they’ll also have a higher chance of falling off their horses and making a fool of themselves.
We also have a command points system, where each player starts with a certain number of command points (let’s say 10) and can spend them to activate special abilities, call for reinforcements, or rally routing units. Command points are replenished at the end of each turn, or when you bribe your opponent with pizza.
That’s a quick overview of the rules, but keep in mind there’s a lot more to it and lot of interesting ideas that can be added as well, like leader figures, random events, and supply system. So don’t forget to bring snacks, because wargaming is a serious business. If you have any questions or anything is unclear, just let me know!
Please note that this is a fictional and humorous way of writing the rules, but it’s important to mention that it’s important for a wargame to be played in a serious and respectful manner.
Now, back to me (Roly).
Here’s a Word document containing the complete conversation we had with ChatGPT to come up with these rules, and including some less light-hearted versions of them.
Whilst the above AI-created wargaming rules are probably not actually playable as written, they are getting pretty close. The whole conversation, including the rules themselves, only took about ten minutes. So spending another hour or too giving ChatGPT some tighter specifications for more specific rules might result in an actual usable ruleset – or certainly one that could be made playable by a few minutes of additional fine-tuning by a human writer.
Though I do note that ChatGPT already writes a much more interesting and funny blog post than I can do!
PS: The header image of Napoleon writing this set of rules on a computer is also AI-created!
An extremely interesting exercise and result! The computer did a fine job of pulling together something ‘new’ from ideas that have been produced before—just like most ‘new’ sets of rules do, hahaha!!
Seriously though, it shows what can be done with fairly ‘clunky’ versions at present putting together something that I’d consider specific and niche. No doubt you’d get more and more with greater specification. Some programming of randomisation could even lead to innovation.
Thanks for posting about this really interesting test.
Regards, James
It certainly was an eye-opener for me, about the possibilities of AI … and also how it could be dangerous too.
Interesting exercise on clunky rules. Thanks for sharing the funny version.
Well, the AI thought it was funny!