Rolled 4, 3 + 5 = 12 (2d6 + 5)>Handwrite monster rough stat block.>Feed it trough chatGPT to point out the flaws like too high or low numbers, overpowered or underpowered mechanics and abilities.>Correct the flaws using both chatGPTs suggestions aswell as comparing it to official stat blocks.>Repeat the process until the stat block is ballanced and usable for it's CR in a real game.Is this morally despicabe of me? Should I stop?
Your game, anon, your rules. Yeah, I'm gonna make fun of you for being too dumb to make your own monster stat block, but hey. It's your game. Do whatever works for you.
>>96607353>morallyI don't give a shit. If it works, it works. If it triggers anyone, then it just filters the spergs I wouldn't want to game with anyways.The only thing that's important to recognize is that AI is kind of abominably shit at game balance or sticking to actual game rules, especially for lesser known systems that aren't D&D, so you have to use your own judgement to curate the output. And that's exactly what you're doing.
moral? I don't knowBut it is retarded. Even with a few months of experience you could identify those issues if you paid attention. A few years in you realize all stat blocks are pretty standard.
>>96607353>Is this morally despicabe of me? Should I stop?What the fuck are you talking about you retard?A thing is morally despicable only if it hurts others. Would you otherwise have paid someone to make the monster for you? No? Then shut up.
>>96607353Heres the thing.You're asking ChatGPT to point out anything under, or overpowered but then double-checking over to make sure it is correct about it bring under or overpowered.You have the critical thinking to assess if it's under or over powered, and you're doing the same amount of work as if you were just checking or balancing your own work. Why bother asking ChatGPT how much acid damage your CR10 Yellow-Dragon homebrew should do with it's breath attack if you already have in your mind "If it says 12d8 or higher, that's way too strong, or if it says lower than 8d8 that's going to be too weak..." So you clearly, already know and have a concept that about 10d8 is the right amount, why ask an AI?Where is the advantage?It feels like asking ChatGPT to summarise a report for you, but also studying the report to make sure the AI got it right. Why bother? What benefit?
>>96607377>And that's exactly what you're doing.Nvm. I replied before reading the OP.Using AI to balance D&D might be fine, because there's so many examples on the web. But D&D is a horribly shittily unbalanced mess anyways, so I think you'll be hurt more by an AI's advice than helped.>>96607353You can keep doing what you're doing, but I don't think it's a good idea, for reasons totally unrelated to your concerns.
>>96607353It's not morally wrong but it's pretty gay, you should be able to judge things for yourself instead of outsourcing your thinking to a machine.
>>96607353>morally despicable In the sense you're wasting opportunities to improve and otherwise engage with your fellows in favour of something simpler and less interesting yes. On the other hand if a llm is providing useful additions to your ideas its unlikely you had much to begin with and you're clearly retarded so your atrophy isn't really a loss. You're likely morally deficient in ways that are of greater consequence.
Chatgpt doesn’t know shit about balance. Stop being a retard.
>>96607353If its complex enough that you cant do it yourself then a LLM is going to be uselessIf its not that complex the LLM is redundant
>>96607353Morality is the idiosyncrasy of decadents having the hidden desire to revenge themselves upon life, anon.Fuck that slave mentality
>>96607353>morallystop being influenced by twitter “artists”
>>96607409What the fuck is a yellow dragon?
>>96607487Gold dragon or homebrew creature
>>96607353how would gpt know how high or low numbers are supposed to be, or what is overpowered or underpowered?
>>96607487
>>96607353That's a Savannah cat, right? Is the armor on to make sure nobody pets it by accident?
>>96607487>>96607543>>96607605yellow dragons are nethack monsters
>>96607353You do not need anyone's validation to do things you want. If it works it works and thats it, move on.
>>96607353>could think of cool monsters>hmm idk I will let the AI do it
>>96607963>Think of cool monster>Design it>Stats are bonkers>Feed trough AI>Adjust stats>Stats not bonkers anymore
>>96607975If you know the stats are wrong then you must already know the right stats are. If you know the stats the AI gave you are right you must already know what the right stats are. So just give it the right stats. I don't this astroturfing. Do you really think this act is going to convince anyone to use AI? Do you get paid more if people use AI? What's your incentive?
>>96608295don't get*
>>96607975I fail to see why you have confidence that the AI has been trained on relevant and balanced material, as opposed to material from five different editions and liberal quantities of horribly unbalanced homebrew.
>>96607975Answer me.
>>96608325No.
>>96608391Not like this.
>>96607975AI suck at math. The stats are going to be bonkers.
>>96607975If you can't determine for yourself whether stats are "bonkers" you shouldn't be running a game.
>>96607353Who cares about morality?All that matters is that the results are going to be predictable or ugly and usually both
>>96607353Yes, you should think for yourself. It's not that hard.
>>96607353>Is it morally despicableThat's going to depend on the system of morals that you subscribe to, but at least some will answer "yes" and some will answer "no".My bigger concern would be that it's retarded: by relying on the AI to do those things, rather than figuring them out yourself, you deny yourself the ability to learn and master the system you're using. Lacking those skills, you'll continue to rely on the AI, probably for an increasing amount of your prep work. But the skills involved in running a campaign are interlinked, so failing to brush up where you can will result in the game overall being worse. And that's all assuming you actually get useful output.
>>96607353AI cannot render judgement, it sucks at it and it's a yes-man style sycophant. TEST IT WITH ACTUAL PEOPLE DUMBFUCK.
>>96607353Stupid is what it is. AI is much better at the initial input 'throwing shit at the wall' stage than it is at the 'fine detail work to be inserted into a system' stage. The damn thing doesn't have the contextual information/understanding to get what something does in a broader context, especially if it doesn't have information on party composition/playstyle/competence.It'd be like expecting a random professor or doctorate to do the same; the issue isn't a lack of neurons firing, its a lack of experience in the material and a lack of contextual information.
If you know the stats are wrong then you must already know what the right stats are. If you know the stats the AI gave you are right you must already know what the right stats are. So just give it the right stats.
>>96607353>Is it morally despicable to use AI tools to assist monster creation?No
Yes
>>96607353>Repeat the process until the stat block is balanced and usable for it's CR in a real game.>Is this morally despicable of me? Should I stop?So long as you're actively trying to balance an encounter, you're in the right in terms of table ethics, so this question is retarded.
>>96607353Yes, I sentence you to a lifetime of horror on elf thread.
>>96607353
>>96616449
>>96616452
>>96607353you'd think a computer would be able to effectively gauge things like averages by category, what with them being literally what we do spreadsheets on, but because of the yes-man directive applied to them, AIs like chatgpt aren't trustworthy even in what's normally a slam-dunk for computer assistance.you said it yourself, you're still double-checking the numbers after checking with the glorified graphing calculator.
>>96607487It's a dragon which, rather than being red, green, blue, black, or white, is yellow.
>>96607353What are you checking the AI's results against?
>>96607353>Is this morally despicabe of me? Should I stop?I can't imagine how. You came up with the monster and statted it out, then just got feedback. Who cares if it's a machine or a person giving you feedback? This sounds exactly what AI *should* be used for.
>>96607353What are the political and social implications of using AI to assist in monster creation? I can only think of the lost work opportunity, and personally, I don't think it should be a given that something like game design be a marketable skill. At most, you could argue the tenuous position that doing something lazily and making a bad product is inherently amoral, but that would assume the premise.
>>96616713It's a reference to a tabletop story, retard.
>>96607353Correcting ChatGPTs creative writing is more work for a worse result than just doing it yourself, plus it decays your brain.It's not morally despicable, you only hurt yourself, but I'll make fun of you, and if I was your player I would leave your table.
>>96619606So what are you confused about?
