So I’ve been using AI to help me with my own little campaign I’ve been trying to build I tried to write it out on my own but im very bad a writing and putting my ideas out there I’ve tried already once and it was a complete train wreck. Now I’m not using AI to write the whole story I give it what I want and it helps me build it and flesh it out more and I put down what I want to come out from it, it helps me build characters and a story I’m combining a module with my homebrew idea.This is just for a private game I’m not sharing this with anyone but I’m also kind of divided because all the problems people have with ai but I just wanted other peoples opinions please go easy On me
also from what I have it’s very similar to what I had in mind I’m just not the greatest when it comes to putting down my ideas and imagination on paper as soon as I get ready to write it my mind goes blank.
just wanted people opinions on if it’s a good idea or if I’m just being stupid
AI /ChatGPT are just tools that could help you if you know how to use it right. I think what you doing is fine, put your main ideas in the bot and use what you read there to help with writing /pacing / scenes or whatever you need.
The advice I give is that you read the results of the Tool, take what you think would be beneficial and ignore the rest, use good "prompts", and try to ask questions in a way you want theme answered, at the end of the day. Its a tool.
As for your private game, ask for feedback from your players, that way you know if you are using it to much or not, so as to not be an ChatGPT runned game instead of you running the game with the ideas, suggestions and text from the tool.
One DM I'm playing with uses ChatGPT (or another) to make the descriptions or session summaries. Sometimes is a little overboard, but most of the time is just an added embelishment to the descriptions or situations.
As a warning, threads on AI typically receive the same basic responses on this forum, often with increasing vitriol and repetitiveness until the thread is ultimately locked. Giving you fair warning on that.
The complaints basically boil down to three camps - (a) the environmental impacts of AI and the ethics of using something that is bad for the planet; (b) the intellectual property impacts of AI, the ethical impacts of using a program trained in a legally grey area; and (c) the fact that AI has flaws in terms of its outputs that make it worse than a talented human DMs.
The first two are valid ethical complaints about AI - but they are ethical complaints you should reach a decision on yourself, rather than crowdsource your ethics from a forum that is extremely polarized on the issue with AI.
The second issue is a bit more of a complex one. In a vacuum and outside of the ethical concerns that I am leaving to you to decide, generative AI does have some utility in D&D - it can be a powerful tool t help a DM with tasks they are not very good at or simply do not enjoy. For example, I use generative AI for NPC art - I am not an artist and never will be talented in that regard. But I can clearly visualize what things look like in my head and can use AI to make up for my artistic shortcomings and help share my visions with the table, giving a sort of shared framework we can all bond over. I also know DMs who use it for generating quick things like bar menus or other busywork, since it can generally be more effective and specific than existing non-AI generators.
I would personally feel uncomfortable using it at the scale you are, and would be skeptical of a DM who does. Great DMing typically requires a DM to know content to a sufficient degree to ad lib wirh that content - when a DM makes that content from scratch, they internalize it more and thus can use it more effectively (this is documented by actual studies - the mind does not internalize AI generated information in the same way).
At the same time, it very certainly results in a worse DM product, but we also have to recognize that there is a longstanding DM shortage, which has resisted decades of mitigating efforts by TSR and Wizards. AI is one of the few tools that might actually help with that, reducing the barrier to entry. A game where the DM heavily relies on AI is not a game I want to play in - but I also have options with a few highly competent DMs and can always just DM myself. For players where the option is between an AI supported DM or no DM and thus no game at all, that math might be very different - and should be made by the DM and players.
I'll be honest, the best way you can DM is to just throw up scenarios, see what the players do in response.
So, first, find a few quests, try to find a narrative to tie them together. Then throw your players at them. I almost always find that something happens during the session that gives me something to develop into more quests. Perhaps they upset a few NPCs, they get bitten by werewolves, they do something that serves as inspiration to develop more quests. Then I just keep following those threads while creating more as the BBEG responds to them.
That allows the campaign to respond to the players while I don't need to rely on AI despite not being the most creative of people.
Rollback Post to RevisionRollBack
If you're not willing or able to to discuss in good faith, then don't be surprised if I don't respond, there are better things in life for me to do than humour you. This signature is that response.
You're not being stupid. What works for you is what works for you. It would help to know a bit more about how much experience you have as a DM. Writing and running a homebrewed campaign if you're still relatively new to DMing can feel a bit like running a marathon while juggling and solving math problems in your head all at the same time. I find that having a simple bullet list of things I can go to if the players take a turn I wasn't expecting: an encounter that could happen anywhere (e.g., bandits that have taken some innocents/travelers hostage, etc.) and/or one or two things that are unexpected (e.g., a pack of hungry wolves shows up in the city). You don't always need to have answers for everything, in fact sometimes the players will give you your answer (as they're trying to figure out reasons as to why wolves would risk coming into the city they might think up several ideas that you could choose from, then build on that one).
My DM experience is nearly 20 years. Recently using an AI to help me make content has been fantastic. I still come up with the ideas, but it helps me flesh them out. The AI can produce a list of NPCs quickly, or just give you some name examples for a village you're improvising.
I wouldn't say you're being stupid, but I'll say I'm of the opinion it's definitely not a good idea.
What you need is practice getting those ideas out on paper (or into a Word doc or whatever). The tried-and-true advice for writers applies here: Write every day. Build the habit of being able to write out what you're brainstorming. Whatever prompts you're feeding into the AI, write them out in your notebook or your app of choice instead and then iterate and expand on it yourself. The more often you exercise doing this yourself the better you'll get at it. You'll also get a better understanding of how the ideas you're envisioning can adapt and change, which is crucial for being ready to adapt to players getting sidetracked and forcing you to improvise. As it stands now you're going in the opposite direction and digging yourself into a rut of cognitive debt.
Also, get to know some other DMs who have been doing this for a while and pick their brains about their creative processes. Don't copy them, of course, but learn what works for them and why it works for them to get a better idea of what can work for you. Try one of the RPG writing workshops at the Storytelling Collective. I learned a lot taking their Write Your First Adventure and Write Your First Encounter courses. And I met a bunch of great people to talk to and learn from.
I think you should practice at it and grow in skill. You do not need GPT.
Rollback Post to RevisionRollBack
He/Him. Loooooooooong time Player. The Dark days of the THAC0 system are behind us.
"Hope is a fire that burns in us all If only an ember, awaiting your call To rise up in triumph should we all unite The spark for change is yours to ignite." Kalandra - The State of the World
If you keep doing it the way you describe, you'll never get any better at doing it on your own. You may actually get worse.
And you will need to build those skills, because DMing is a heavily improvisational activity. There's no way to avoid it, and you can't use ChatGPT in the moment.
We were all bad at this once upon a time. We got better by doing it.
I wouldn't say you're being stupid, but I'll say I'm of the opinion it's definitely not a good idea.
What you need is practice getting those ideas out on paper (or into a Word doc or whatever). The tried-and-true advice for writers applies here: Write every day. Build the habit of being able to write out what you're brainstorming. Whatever prompts you're feeding into the AI, write them out in your notebook or your app of choice instead and then iterate and expand on it yourself. The more often you exercise doing this yourself the better you'll get at it. You'll also get a better understanding of how the ideas you're envisioning can adapt and change, which is crucial for being ready to adapt to players getting sidetracked and forcing you to improvise. As it stands now you're going in the opposite direction and digging yourself into a rut of cognitive debt.
Also, get to know some other DMs who have been doing this for a while and pick their brains about their creative processes. Don't copy them, of course, but learn what works for them and why it works for them to get a better idea of what can work for you. Try one of the RPG writing workshops at the Storytelling Collective. I learned a lot taking their Write Your First Adventure and Write Your First Encounter courses. And I met a bunch of great people to talk to and learn from.
And if you’re worried about your players throwing you a curve ball that derails the whole session (it’s bound to happen eventually) look into taking improv classes. Trust me, learning when to “yes, and…” a situation can be a life saver
I use AI in my current games. Strongly worded prompts and prompt templates to ensure I get what I want out of the AI. Why? Because my prompts include language around variability and surprise (to me as the DM). It helps me shake up old patterns.
It's a prep tool. Table use slows me down and isn't helpful.
Also, for anyone who's curious but want to remain ethical - look into "Plotto" as a help. It's a writer's plot generating algorithm that's in the public domain (it's 100 is years old). AI can use that and generate situations and narratives easy.
Also, for anyone who's curious but want to remain ethical - look into "Plotto" as a help. It's a writer's plot generating algorithm that's in the public domain (it's 100 is years old). AI can use that and generate situations and narratives easy.
Just to address this point... unless you are training your own AI, the poison is already in the well, as it were. You can feed it a document and tell it to use only that, but the current LLMs are all already trained on copyrighted material they didn't pay for. Asking it not to use that in your prompt doesn't stop the fact that it's already happened.
Last night I had a brief but productive D&D adventure writing session that serves as a good example of what I meant about writing out ideas and developing them.
Earlier in the day yesterday I scribbled out a half-sentence brainstorming idea that seemed interesting but I had no idea what to do with. Later on last night I did a 30-minute writing session in which I started writing down questions about that brainstorming idea, and as I asked myself more questions it turned into writing out plot hooks, characters, motivations, branching story paths, and the bones of a boss fight, which all made up a rough outline for a one-shot adventure that draws inspiration from The Mummy (1999), Maverick, Hot Tub Time Machine, and Final Destination. It needs cleaning up, sure, but for a half-hour writing sprint, it ain't half bad.
This kind of creative exercise pays dividends by building habits, developing skills, and accumulating a bank of developed encounter and adventure ideas to tap into.
I use AI in my current games. Strongly worded prompts and prompt templates to ensure I get what I want out of the AI. Why? Because my prompts include language around variability and surprise (to me as the DM). It helps me shake up old patterns.
It's a prep tool. Table use slows me down and isn't helpful.
Also, for anyone who's curious but want to remain ethical - look into "Plotto" as a help. It's a writer's plot generating algorithm that's in the public domain (it's 100 is years old). AI can use that and generate situations and narratives easy.
I agree with this. The AI tools can be helpful with grinding out some of the details that can be taxing after doing it for long enough. Things like NPC names, the personal history of some minor NPCs, and details about random groups are things that I'll tap into AI to smooth some edges.
I think of AI as being a usually less predictable series of random charts. I have 2nd and 3/3.5 edition books that have chats for everything that AI offers, but working full time with two kids doing sports/music means that I have less time and energy. The ease of just asking for details about the local town sheriff posse from an AI is much quicker and less predictable than me deciding which chart I wanna pull, finding the chart in the book, rolling on multiple chats to see what it suggest, and then customizing if to fit the setting and narrative. I mean you should be to do the last storm
Also, for anyone who's curious but want to remain ethical - look into "Plotto" as a help. It's a writer's plot generating algorithm that's in the public domain (it's 100 is years old). AI can use that and generate situations and narratives easy.
Just to address this point... unless you are training your own AI, the poison is already in the well, as it were. You can feed it a document and tell it to use only that, but the current LLMs are all already trained on copyrighted material they didn't pay for. Asking it not to use that in your prompt doesn't stop the fact that it's already happened.
You can obtain the text of Plotto for free online. You can also order a printed copy (I did). Then you can use PLOTTO's system absent any AI to generate plots. That's why I called it an ethical option. You need zero AI to use it.
Also, for anyone who's curious but want to remain ethical - look into "Plotto" as a help. It's a writer's plot generating algorithm that's in the public domain (it's 100 is years old). AI can use that and generate situations and narratives easy.
Just to address this point... unless you are training your own AI, the poison is already in the well, as it were. You can feed it a document and tell it to use only that, but the current LLMs are all already trained on copyrighted material they didn't pay for. Asking it not to use that in your prompt doesn't stop the fact that it's already happened.
You can obtain the text of Plotto for free online. You can also order a printed copy (I did). Then you can use PLOTTO's system absent any AI to generate plots. That's why I called it an ethical option. You need zero AI to use it.
I'm glad you can. It's just that your last sentence in that quote mentioned using AI with it.
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
So I’ve been using AI to help me with my own little campaign I’ve been trying to build I tried to write it out on my own but im very bad a writing and putting my ideas out there I’ve tried already once and it was a complete train wreck. Now I’m not using AI to write the whole story I give it what I want and it helps me build it and flesh it out more and I put down what I want to come out from it, it helps me build characters and a story I’m combining a module with my homebrew idea.This is just for a private game I’m not sharing this with anyone but I’m also kind of divided because all the problems people have with ai but I just wanted other peoples opinions please go easy On me
also from what I have it’s very similar to what I had in mind I’m just not the greatest when it comes to putting down my ideas and imagination on paper as soon as I get ready to write it my mind goes blank.
just wanted people opinions on if it’s a good idea or if I’m just being stupid
AI /ChatGPT are just tools that could help you if you know how to use it right.
I think what you doing is fine, put your main ideas in the bot and use what you read there to help with writing /pacing / scenes or whatever you need.
The advice I give is that you read the results of the Tool, take what you think would be beneficial and ignore the rest, use good "prompts", and try to ask questions in a way you want theme answered, at the end of the day. Its a tool.
As for your private game, ask for feedback from your players, that way you know if you are using it to much or not, so as to not be an ChatGPT runned game instead of you running the game with the ideas, suggestions and text from the tool.
One DM I'm playing with uses ChatGPT (or another) to make the descriptions or session summaries. Sometimes is a little overboard, but most of the time is just an added embelishment to the descriptions or situations.
As a warning, threads on AI typically receive the same basic responses on this forum, often with increasing vitriol and repetitiveness until the thread is ultimately locked. Giving you fair warning on that.
The complaints basically boil down to three camps - (a) the environmental impacts of AI and the ethics of using something that is bad for the planet; (b) the intellectual property impacts of AI, the ethical impacts of using a program trained in a legally grey area; and (c) the fact that AI has flaws in terms of its outputs that make it worse than a talented human DMs.
The first two are valid ethical complaints about AI - but they are ethical complaints you should reach a decision on yourself, rather than crowdsource your ethics from a forum that is extremely polarized on the issue with AI.
The second issue is a bit more of a complex one. In a vacuum and outside of the ethical concerns that I am leaving to you to decide, generative AI does have some utility in D&D - it can be a powerful tool t help a DM with tasks they are not very good at or simply do not enjoy. For example, I use generative AI for NPC art - I am not an artist and never will be talented in that regard. But I can clearly visualize what things look like in my head and can use AI to make up for my artistic shortcomings and help share my visions with the table, giving a sort of shared framework we can all bond over. I also know DMs who use it for generating quick things like bar menus or other busywork, since it can generally be more effective and specific than existing non-AI generators.
I would personally feel uncomfortable using it at the scale you are, and would be skeptical of a DM who does. Great DMing typically requires a DM to know content to a sufficient degree to ad lib wirh that content - when a DM makes that content from scratch, they internalize it more and thus can use it more effectively (this is documented by actual studies - the mind does not internalize AI generated information in the same way).
At the same time, it very certainly results in a worse DM product, but we also have to recognize that there is a longstanding DM shortage, which has resisted decades of mitigating efforts by TSR and Wizards. AI is one of the few tools that might actually help with that, reducing the barrier to entry. A game where the DM heavily relies on AI is not a game I want to play in - but I also have options with a few highly competent DMs and can always just DM myself. For players where the option is between an AI supported DM or no DM and thus no game at all, that math might be very different - and should be made by the DM and players.
I'll be honest, the best way you can DM is to just throw up scenarios, see what the players do in response.
So, first, find a few quests, try to find a narrative to tie them together. Then throw your players at them. I almost always find that something happens during the session that gives me something to develop into more quests. Perhaps they upset a few NPCs, they get bitten by werewolves, they do something that serves as inspiration to develop more quests. Then I just keep following those threads while creating more as the BBEG responds to them.
That allows the campaign to respond to the players while I don't need to rely on AI despite not being the most creative of people.
If you're not willing or able to to discuss in good faith, then don't be surprised if I don't respond, there are better things in life for me to do than humour you. This signature is that response.
You're not being stupid. What works for you is what works for you. It would help to know a bit more about how much experience you have as a DM. Writing and running a homebrewed campaign if you're still relatively new to DMing can feel a bit like running a marathon while juggling and solving math problems in your head all at the same time. I find that having a simple bullet list of things I can go to if the players take a turn I wasn't expecting: an encounter that could happen anywhere (e.g., bandits that have taken some innocents/travelers hostage, etc.) and/or one or two things that are unexpected (e.g., a pack of hungry wolves shows up in the city). You don't always need to have answers for everything, in fact sometimes the players will give you your answer (as they're trying to figure out reasons as to why wolves would risk coming into the city they might think up several ideas that you could choose from, then build on that one).
My DM experience is nearly 20 years. Recently using an AI to help me make content has been fantastic. I still come up with the ideas, but it helps me flesh them out. The AI can produce a list of NPCs quickly, or just give you some name examples for a village you're improvising.
I wouldn't say you're being stupid, but I'll say I'm of the opinion it's definitely not a good idea.
What you need is practice getting those ideas out on paper (or into a Word doc or whatever). The tried-and-true advice for writers applies here: Write every day. Build the habit of being able to write out what you're brainstorming. Whatever prompts you're feeding into the AI, write them out in your notebook or your app of choice instead and then iterate and expand on it yourself. The more often you exercise doing this yourself the better you'll get at it. You'll also get a better understanding of how the ideas you're envisioning can adapt and change, which is crucial for being ready to adapt to players getting sidetracked and forcing you to improvise. As it stands now you're going in the opposite direction and digging yourself into a rut of cognitive debt.
Also, get to know some other DMs who have been doing this for a while and pick their brains about their creative processes. Don't copy them, of course, but learn what works for them and why it works for them to get a better idea of what can work for you. Try one of the RPG writing workshops at the Storytelling Collective. I learned a lot taking their Write Your First Adventure and Write Your First Encounter courses. And I met a bunch of great people to talk to and learn from.
I think you should practice at it and grow in skill. You do not need GPT.
He/Him. Loooooooooong time Player.
The Dark days of the THAC0 system are behind us.
"Hope is a fire that burns in us all If only an ember, awaiting your call
To rise up in triumph should we all unite
The spark for change is yours to ignite."
Kalandra - The State of the World
If you keep doing it the way you describe, you'll never get any better at doing it on your own. You may actually get worse.
And you will need to build those skills, because DMing is a heavily improvisational activity. There's no way to avoid it, and you can't use ChatGPT in the moment.
We were all bad at this once upon a time. We got better by doing it.
And if you’re worried about your players throwing you a curve ball that derails the whole session (it’s bound to happen eventually) look into taking improv classes. Trust me, learning when to “yes, and…” a situation can be a life saver
I've been DM'ing all on my own since 1983.
I use AI in my current games. Strongly worded prompts and prompt templates to ensure I get what I want out of the AI. Why? Because my prompts include language around variability and surprise (to me as the DM). It helps me shake up old patterns.
It's a prep tool. Table use slows me down and isn't helpful.
Also, for anyone who's curious but want to remain ethical - look into "Plotto" as a help. It's a writer's plot generating algorithm that's in the public domain (it's 100 is years old). AI can use that and generate situations and narratives easy.
Just to address this point... unless you are training your own AI, the poison is already in the well, as it were. You can feed it a document and tell it to use only that, but the current LLMs are all already trained on copyrighted material they didn't pay for. Asking it not to use that in your prompt doesn't stop the fact that it's already happened.
Last night I had a brief but productive D&D adventure writing session that serves as a good example of what I meant about writing out ideas and developing them.
Earlier in the day yesterday I scribbled out a half-sentence brainstorming idea that seemed interesting but I had no idea what to do with. Later on last night I did a 30-minute writing session in which I started writing down questions about that brainstorming idea, and as I asked myself more questions it turned into writing out plot hooks, characters, motivations, branching story paths, and the bones of a boss fight, which all made up a rough outline for a one-shot adventure that draws inspiration from The Mummy (1999), Maverick, Hot Tub Time Machine, and Final Destination. It needs cleaning up, sure, but for a half-hour writing sprint, it ain't half bad.
This kind of creative exercise pays dividends by building habits, developing skills, and accumulating a bank of developed encounter and adventure ideas to tap into.
I agree with this. The AI tools can be helpful with grinding out some of the details that can be taxing after doing it for long enough. Things like NPC names, the personal history of some minor NPCs, and details about random groups are things that I'll tap into AI to smooth some edges.
I think of AI as being a usually less predictable series of random charts. I have 2nd and 3/3.5 edition books that have chats for everything that AI offers, but working full time with two kids doing sports/music means that I have less time and energy. The ease of just asking for details about the local town sheriff posse from an AI is much quicker and less predictable than me deciding which chart I wanna pull, finding the chart in the book, rolling on multiple chats to see what it suggest, and then customizing if to fit the setting and narrative. I mean you should be to do the last storm
You can obtain the text of Plotto for free online. You can also order a printed copy (I did). Then you can use PLOTTO's system absent any AI to generate plots. That's why I called it an ethical option. You need zero AI to use it.
I'm glad you can. It's just that your last sentence in that quote mentioned using AI with it.