"Are AI-Generated Rulesets Among Us?" Topic

How worried are you about AI-generated wargaming rulesets and related products?

How would you recognize an AI-generated product?

I am not at all worried, and if it were a good AI-generated rues set, I wouldn't be able to recognize it.

Ask me how irked I am about rules sets written by suits to force franchise wargamers to buy or modify armies.

How irked are you about rules sets written by suits to force franchise wargamers to buy or modify armies?

Lol

Seriously. They're a discredit to historical miniatures gaming, and not even fantasy and SF gamers deserve to be treated in such a manner. Some of the larger wargame companies rely on customer loyalty while doing very little to deserve that loyalty.

Thank you for asking.

Most likely any AI system will evaluate the rules systems available and generate something very similar to the ones available today.

However, if you can customize your inquiry directing it towards something the current rule system does not include, you may come up with something new.

I inquired on ChatGPT as to what the best WWII miniature rule sets were.

Answer: CoC, FoW, and Bolt Action

Sorry UshCha.

Wolfhag

I think AI rule sets will be good at getting rid of errors in spelling, grammar, cross referenced values/numbers.

AI will probably also pick up on common wargamer wrong grammar. eg. mechanic/mechanism faction/nation. Although it can probably instructed to "get it wrong" deliberately?

Overall AI should be good. It can maybe take the author's ideas and get them to work as a whole?

AI will mainly be derivative of the rules it reads (it can read them all) but not able to create new ideas as ideas can be totally original/invented.

Just thoughts.

martin

Martin, I wish even that limited possibility were true. But AI cannot create a consistent set of rules; the text model just puts words in a grammatically correct order which reflects everything itt used to teach itself without any understanding. I'm fairly sure that were I to ask AI to write a 'set of rules' that at first scan seems like a set of rules, but which on even a cursory examination would prove to be incomplete and inconsistent. So Bill, unless you know that this is happening now, I'll take the 'this is ridiculous' option.

As a thought experiment I asked an AI to write WW2 rules and it basically wrote Bolt Action. No slam on Bolt Action, it was probably the most frequently mentioned rules online.

I don't know, if the rules are good why should I care if AI made them.

I have great hopes for AI--but largely for human-guided AI fiction. ("Write a novella set on Mars in the style of Leigh Brackett or Edgar Rice Burroughs.")

For rules, I think I've read a number which reflect everything the author used to teach himself without any understanding, and they were all written by people. To me, scenario design has elements of art, but rules writing is pretty much mechanical. Once you've made the input decisions--ground scale, time scale, basing, casualty removal, morale, randomizing method and weapons ranges and effects--you're so close to the final product that you stand in more need of SpellCheck and GrammarCheck than of anything to generate text.

But if software could generate rules I can't distinguish from a human product, it passes the Turing test, miniatures version. Why should I care?

Concur, don't care.

AI is a toolset. I don't care what toolset was used. I care about the quality of the output.

I am, however, very concerned about how rules force people to buy and modify armies. Do they brandish weapons and give orders? Threaten loved ones' safety? Blackmail wargamers with exposure of their deep, dark secrets? Maybe the rules are blocking access to critical medical care? A disturbing dystopia, indeed…

I wouldn't mind an AI opponent as long as it was easy to use and not impossible to occasionally get the better of.

Above is obviously not AI generated grammar.

"Solo play" rules are (or, can be) AI opponents:

link

This is opposed to one of my favourite non-player opponents is the Pandemic board game. While "AI" is not actually a thing, one of the general expectations is that the automated decision making is different in response to what the human players do. In Pandemic the outcomes of the automated decisions are affected by both sides, but the decisions themselves are not.

Fair point, eto. Would you have been happier had I written "rules written by suits which invalidate existing armies and compel persons playing the franchise to purchase additional forces to play with the other gamers of that franchise?"

But like offering a unique army/scale combination and never getting the line to a wargameable point, it shows a contempt for the customer seldom works out well for either party over the long term. To the extent there is a "wargaming community" that sort of behavior undercuts it.

I would prefer "too many people in my area who abandon something I wanted to continue working on when something new comes out."

It is consumer-driven behaviour, not producer-driven. Nothing compels them other than their own desires.

What we are currently calling AI these days can't "write" anything. All it can do is plagiarize already existing works. Original concepts and original text are beyond the capability of an AI; derivative product, however, it can do. Whether that product will have a logical internal consistency or be "fun" is another matter.

Which doesn't mean that an AI might not generate a viable game, especially when subjected to the interpretation of a human motivated to "make it work,"— sort of like the original Dungeons & Dragons game. Much of that relied on imaginative interpretation by the players (and especially the DM) to handle all the situations not covered by the rules— including, somewhat surprisingly, any methods for combat procedure. So an open-ended game which people can "play" because their desire to do so causes them to fill in all the missing bits and correct the inconsistencies… yeah, that an AI can do. Though why one would bother to have one do so I can't fathom.

Real AI— true "Artificial Intelligence" is not what we are dealing with today. What is AI today is a simply a high-speed search and agglomerate system which has the ability to correct its approach if told to do so. But it is not true intelligence.

Like a number of people I asked an AI generator to write a set of rules.

It went out and gabbed sections from whatever it found on-line. The sections didn't hang together and much of it was simply gibberish.

There was no game to be had at the end of the process as too much was missing.

You could even tell what rules sets it stole from.

Having said that; after several repeats, it sometimes threw together disparate ideas for a particular rule that might be worth developing further.

Regards

David F Brown

Actually, eto, your statement would not be true for me. I've never fallen down the franchise gaming hole. I am merely sympathetic with people who thought they bought rules and an army, then discovered they were only rented.

Keep in mind that this is the worst AI will ever be, it will only keep improving and likely will improve quickly. It may not be able to crank out a full-fledged rules set or game system yet but even generating a decent first-pass draft would speed things up considerably. There would still be a need for human editing and fact-checking so I don't think game designers would be out of a job, but AI might make them much more productive.

I'm with Inch High Guy here. AI is only going to improve and will be one more item in the creator's arsenal.

Inch High Guy and Dentatus +1

sympathetic with people who thought they bought rules and an army, then discovered they were only rented.

I find it hard to sympathize for someone who feels bad because they chose to buy more rules and not play the ones they have.

As far as "rented" goes, I still have my copy of original Space Hulk. No GW minions have broken in to my house and taken it. I have yet to receive a C&D letter for playing it over the last four or five decades.

Wolfhag, Do not feel sorry, were AI to praise my rules I would be worried, praise from the wrong sort is worse than anything .

People seem to be all Verklempt over AI using machine learning algorithms to produce derivative works.

BUT, if I were to write a ruleset with what I consider the "best of" mechanics, that's fine.

Well, yes, eto--but now I'll have to turn you in to GW for the reward.

More seriously, how can AI-generated rules sets be perceived as some sort of threat, hiding among us? There is no limit on number of rules which can be published in a year, so they're not taking anyone's spot. If they're bad rules, hopefully fewer people will buy them, and Heaven knows if bad rules could sink wargaming, we'd have been at the bottom of the sea decades ago. If the rules are clearly written, easy to play and produce historical results, surely they would be of benefit to miniatures gaming, no matter how they were arrived at?

AI will do this, AI will wreck that, AI will solve those.

Why not stick to the Celts greatest fear 'the sky will fall tomorrow' – just as logical. However, the celts had some reason to be fearful as they knew very little about how the sky worked. You could go and find out easily and realise what it really is and what it can and can't do.

UshCha,
LOL, I know that would get you to respond and that you don't care.

Here is a ChatGPT inquiry:
Me: write me some rules that would simulate small arms suppression on an infantry platoon in WWII

ChatGPT:

Simulating small arms suppression on an infantry platoon in World War II involves considering various factors that contribute to the chaos and danger of the battlefield. Here are some rules to simulate small arms suppression:

Suppressive Fire Mechanics:
Designate a specific area or sector on the battlefield as the target for suppressive fire.
Establish a set rate of fire and ammunition expenditure for the suppressing unit.
Determine the effectiveness of suppressive fire in limiting the movement and actions of the targeted infantry platoon.

Cover and Concealment:
Define different levels of cover (e.g., trenches, sandbags, buildings) and concealment (e.g., vegetation) on the battlefield.
Establish rules for the impact of suppressive fire on the effectiveness of cover and concealment.

Fear and Morale Mechanics:
Introduce a morale system for infantry units that takes into account the intensity and proximity of suppressive fire.
Develop rules for the potential impact of fear and stress on the ability of the infantry platoon to execute commands and maintain cohesion.

Movement and Actions Under Suppression:
Specify penalties for movement and actions taken by the suppressed infantry platoon, such as reduced accuracy, slower movement, or limited actions per turn.
Allow for the possibility of pinned or immobilized units as a result of intense suppressive fire.

Command and Control:
Implement rules for the command structure within the infantry platoon, affecting their ability to coordinate and respond to suppressive fire.
Consider the impact of leadership and communication breakdowns on the effectiveness of the platoon under suppression.

Resilience and Training:
Introduce factors such as the experience level and training of the infantry platoon to determine their resilience under suppressive fire.
Account for the potential of elite or well-trained units to better withstand and respond to suppression.

Ammunition and Resource Management:
Track the ammunition levels of the suppressing unit, and establish rules for reloading or resupply.
Consider the impact of ammunition shortages on the sustainability of suppressive fire.

Suppression Duration and End Conditions:
Set a timeframe for the duration of suppressive fire and establish conditions under which the suppression is lifted (e.g., time, achieving specific objectives, or breaking the morale of the suppressing unit).

These rules provide a foundation for simulating small arms suppression in a WWII infantry platoon setting. Tailor them to fit the specific rules and mechanics of your tabletop or simulation game.

Notice it did not write anything, it just gave suggestions. It appears to be just what evilgong said.

Right now AI appears to be plagiarizing search engines but will improve, especially if you can customize it.

Wolfhag

Yup like most one page rules unplayable no change there then.

So, what ChatGPT actually did was generate an assignment. Its output looks very much like the brief I mightt hand to a game designer hired to write my WW2 small action rules for me.

There are no actual rules or mechanisms there.

Have purchased, read, and tried to play some truly horrendous human generated war-games rules, and even written some myself, some rules look great in theory on paper but utterly fall apart when the minis hit the table, I do not see AI as much of threat yet.

If and when AI can generate good enough rules, faster and cheaper than a human then I will worry. This is already happening in the field of illustration. AI is faster, cheaper, and often good enough. Some companies are now actually hiring artists to train the AIs that have replaced artists to do a better job of replacing artists. It's cheaper that way.

As for the rules themselves, I want a playable, enjoyable game. For historical subjects this includes something that recreates the history. For fantasy and science fiction it includes something that matches the fiction. If an AI can produce rules that give an enjoyable game I would play them. If I human writes a set of rules that give an enjoyable game I would play them. If the game is not enjoyable to me, I won't play it. I imagine I am like a lot of warmers in that respect. We may differ on what makes a game "enjoyable" or "satisfying" but if it isn't we will likely be looking for something better suited. For all the people who enjoy Bolt Action and Flames of War I don't imagine they would satisfy UshCha!

You
Make me a set of RPG rules called The Information Super Highway Exponential Brain Game

ChatGPT
copy

Introduction: Welcome to the Information Super Highway Exponential Brain Game (ISHEBG), a role-playing game that enhances your cognitive abilities while exploring the vast expanse of the digital world. In this game, players will navigate through the challenges of the Information Super Highway, solving puzzles, completing quests, and acquiring knowledge along the way. Get ready to exercise your brainpower and embark on an epic digital adventure!

Core Concepts:

Character Creation:

Players create a character by selecting a Class (such as Hacker, Data Analyst, Technomancer) and assigning attributes (Intelligence, Agility, Charisma, etc.).
Each class has unique abilities and skills related to their area of expertise.
Attributes:

Intelligence: Determines problem-solving abilities and knowledge acquisition.
Agility: Affects reaction time and quick thinking.
Charisma: Influences negotiation skills and interaction with other digital entities.
Gameplay:

Players navigate the Information Super Highway, which is divided into different interconnected domains.
Each domain poses puzzles, challenges, or quests that need to be completed to progress.
Encounters:

Encounters can be with digital entities, obstacles, or puzzles.
Players must use their character's attributes and skills to overcome these challenges.
Success in encounters may reward players with experience points, items, or information.
Levelling Up:

By earning experience points, characters can level up.
Leveling up improves attributes and unlocks new abilities.
Abilities and Skills:

Characters gain abilities based on their class, such as hacking, data analysis, or virtual combat.
Skills can be acquired through quest completion or learning from digital entities along the Information Super Highway.
Knowledge Acquisition:

Characters can gain knowledge by interacting with specific digital entities or completing educational quests.
Knowledge improves problem-solving abilities and grants access to previously inaccessible domains.
Cybernetic Enhancements:

Characters can acquire cybernetic enhancements that provide special abilities or boost attributes.
Enhancements may be purchased, found, or rewarded during gameplay.
Social Interaction:

Players can team up with other characters or interact with NPCs.
Cooperative gameplay allows for collaborative problem-solving and quest completion.
Endgame:

The ultimate goal is to uncover the reason behind the creation of the Information Super Highway and solve the mysteries found within.
Players may encounter a final boss or brain-teasing challenges to test their accumulated knowledge and skills.
Remember, the Information Super Highway Exponential Brain Game is not just about exploration; it's also about expanding your mental abilities. Sharpen your intellect, adapt to new challenges, and conquer the digital frontier!

Note: Feel free to modify or expand upon these rules to suit your desired gameplay, mechanics, and story for The Information Super Highway Exponential Brain Game.)

I'm mildly obsessed with uncovering the reason behind the creation of the Information Super Highway and the mysteries found within! I may have to devise my own mechanics but it, thank you Chat GPT for laying out the goal of ISHEGP! A game I coined with my buddy at 4AM August 2nd, 1994, Canberra Australia. Status:unemployed, unmarried and virgins.

Status 2024: unemployed, unmarried, virgin. Have been employed before those, in the past. between 1994 and 2024, ie FYI I have been employed for at least 7 years full time in those times. Thanks.

Just for fun I asked ChatGPT to write a set of gladiator rules. This is what it came up with

Gladiator Combat Wargame Rules

Objective: The goal of the game is to eliminate the opposing gladiator(s) or achieve specific victory conditions.

Setup:
1. Each player assembles a gladiator warband using miniatures. A standard warband consists of 3-5 gladiators.
2. Choose a tabletop arena with terrain features like pillars, pits, and obstacles.
3. Determine deployment zones on opposite ends of the tabletop.

Turn Sequence:
1. Initiative Phase:
• Roll a six-sided die (D6) for each player. The player with the highest roll goes first.
• In case of a tie, reroll.
2. Activation Phase:
• The player with the initiative activates one of their gladiators, followed by the opponent activating one of theirs.
• Repeat until all gladiators have been activated.
3. Gladiator Activation:
• During a gladiator's activation, the player can perform two actions: Move, Attack, or Special Action.
• Move: Gladiators can move up to their movement value in inches.
• Attack: Roll a D6 to hit. The target's armor value is subtracted from the roll. If the result is equal to or higher than the target's armor, it's a successful hit.
• Damage is determined by rolling another D6 and consulting a weapon chart. Apply modifiers based on the weapon used.
• Special Action: Some gladiators may have special abilities or actions unique to their class. Use these strategically.
4. Morale Phase:
• Roll a D6 for each gladiator that suffered casualties or was engaged in combat during the turn.
• If the roll is equal to or less than the gladiator's morale value, they pass the morale test. Otherwise, they may suffer penalties in the next turn.
5. End of Turn:
• Check victory conditions. If met, the game ends. Otherwise, proceed to the next turn.

Optional Rules:
1. Terrain Effects: Assign specific effects to terrain features, like defensive bonuses for pillars or danger for pits.
2. Crowd Favor: Introduce a crowd favor mechanic, where successful or stylish attacks earn favor points. Players can spend these points for bonuses or special actions.
3. Campaign Mode: Create a campaign system where gladiators gain experience, learn new skills, and acquire better equipment between battles.
Remember to adapt and balance the rules based on your preferences and the dynamics you want in your gladiator combat tabletop wargame.

1. Thracian Retiarius (Net-and-Trident Gladiators):
• Movement: 6 inches
• Armor: 3
• Morale: 8
• Weapon: Trident (D6 damage)
• Special: Net – Instead of making a regular attack, the Retiarius can attempt to ensnare an opponent with a net. Roll a D6; on a 4 or higher, the opponent is entangled, reducing their movement by half until they break free.

2. Secutor (Sword-and-Shield Gladiators):
• Movement: 5 inches
• Armor: 4
• Morale: 7
• Weapon: Gladius (D6 damage)
• Shield: +1 Armor when defending

3. Murmillo (Heavy Gladiator with Large Shield):
• Movement: 4 inches
• Armor: 5
• Morale: 7
• Weapon: Scutum (D6 damage)
• Special: Gladiatorial Stance – The Murmillo gains +1 to hit and +1 to armor while standing still and not attacking.

4. Retiarius (Trident-and-Net Gladiator):
• Movement: 6 inches
• Armor: 3
• Morale: 8
• Weapon: Trident (D6 damage)
• Special: Net – Same as Thracian Retiarius

5. Dimachaerus (Dual-Wield Gladiator):
• Movement: 5 inches
• Armor: 3
• Morale: 7
• Weapons: Dual Gladii (2x D6 damage)
• Special: Dual Wield – Dimachaerus can make two attacks in a single activation, each at a -1 to hit penalty.

6. Provocator (Long-Reach Gladiator with Shield):
• Movement: 5 inches
• Armor: 4
• Morale: 7
• Weapon: Spear (D6 damage)
• Shield: +1 Armor when defending
• Special: Reach – Provocator has a 2-inch reach for attacks.

These are just basic guidelines, and you can further customize the statistics based on playtesting and personal preferences. Consider adding unique abilities, strengths, or weaknesses for each gladiator type to enhance the tactical depth of the game.

That looks very familiar, possibly straight plagiarism from an existing set ?

Probably, although I have no idea which set. Anyone recognise them?