| 1 person marked this as a favorite. |
So ChatGPT has been all over the news lately and I'm wondering if anybody has any experience using it as a Gamemastering aid.
Obviously, it's nowhere near the point of replacing a GM altogether, but when I typed in "What are some original story ideas for a Dungeons and Dragons campaign?" (I wasn't sure if Pathfinder was prominent enough in the training data for it to know what that was). The machine then gave me these:
2. The players are hired by a wealthy merchant to retrieve a rare and valuable artifact from a long-abandoned temple, only to discover that the temple is haunted by the spirits of those who died trying to claim the artifact.
3. A plague has swept across the land, and the players must find a cure before it consumes the entire kingdom.
4. The players are hired by a group of dwarves to help them reclaim their ancestral homeland, which has been taken over by a group of orcs.
5. A powerful sorceress has seized control of a kingdom, using her magic to keep the populace in line. The players must find a way to defeat her and restore freedom to the people.
6. The players are recruited by a group of rebels fighting against an oppressive government, and must use their skills and wits to bring down the corrupt rulers and bring about a new era of freedom.
7. A group of powerful dragons has declared war on the mortal races, and the players must find a way to stop them before it's too late.
8. A long-dormant volcano has suddenly erupted, unleashing a horde of monsters that threaten to overrun the surrounding lands. The players must brave the dangers of the volcano and put a stop to the threat.
They're all pretty standard, though number 2 seems more like an adventure than a campaign (though it could become a campaign with some improvisation and further development). I rolled a d8, got a 7, and typed in "In number 7, why did the group of powerful dragons declare war on the mortal races?" and got this response:
1. The dragons feel that the mortal races have become too powerful and pose a threat to their dominance. They may see war as the only way to protect their own interests.
2. The dragons may have grown tired of living in hiding and have decided to reclaim their rightful place as the dominant species on the planet.
3. The dragons may have been provoked by the actions of the mortal races, such as encroaching on their territory or stealing their treasure.
4. The dragons may be under the influence of a powerful force, such as a demon or an evil deity, that has convinced them to wage war on the mortal races.
5. The dragons may be seeking revenge for some perceived wrong that the mortal races have inflicted upon them in the past.
Ultimately, the reason for the dragons' declaration of war could be a key plot point in the campaign, and the players may need to uncover the reason in order to find a way to stop the conflict and bring about a peaceful resolution.
I decided that while some of these reasons were not mutually contradictory and could be woven together to make the story more engaging, I was most interested in trying to pin the AI down on #4 and see how good its world building skills were. So I asked "In number 4, how did the evil deity convince the dragons to wage war?"
1. The evil deity may have promised the dragons great power and rewards in exchange for their loyalty and service.
2. The evil deity may have appealed to the dragons' pride and ego, convincing them that they were the rightful rulers of the world and that the mortal races were nothing but inferior beings.
3. The evil deity may have used trickery and deceit to manipulate the dragons into believing that the mortal races were a threat that needed to be eliminated.
4. The evil deity may have used its power to control the dragons' minds, forcing them to do its bidding.
5. The evil deity may have used a combination of these methods to convince the dragons to wage war on the mortal races.
Ultimately, the specific methods that the evil deity used to convince the dragons to wage war could be an interesting plot point for the campaign, and the players may need to uncover the truth and find a way to break the deity's hold over the dragons in order to bring about a peaceful resolution.
So number 5 is kind of a cop-out, but at least the machine is smart enough to realize that these methods are not mutually exclusive. Now, a story needs a cast of characters besides the villains and the PCs, so my next question was: "Other than the players, who resists the dragons most fiercely?"
The AI had a bit of a hiccup here, accidentally listing as item number 2 out of 5 a group that would ally with the dragons and why, so I downvoted that response and asked for another.
1. The mortal races: Depending on the setting and the nature of the conflict, it's possible that the mortal races (such as humans, dwarves, elves, etc.) may band together to resist the dragons' aggression. They may have their own armies and leaders, or they may rely on the players to help them fight back against the dragons.
2. Other powerful creatures: It's possible that other powerful creatures, such as dragons of different alignments, powerful wizards, or ancient beings, may decide to resist the dragons and join the fight against them. These creatures may have their own motivations for opposing the dragons, and they may provide valuable support to the players and the mortal races in their efforts to stop the war.
3. Religious or ideological groups: It's possible that certain religious or ideological groups may take up arms against the dragons, either out of a sense of duty or because they see the dragons' actions as a threat to their way of life. These groups may be willing to make great sacrifices to stop the dragons and protect their beliefs.
4. Rogue adventurers: It's possible that some adventurers may decide to resist the dragons on their own, either for personal gain or because they see the dragons' actions as evil and want to do something about it. These adventurers may be lone wolves or may work in small groups, and they could provide valuable support to the players in their efforts to stop the war.
My next question was, "How do the dragons initiate hostilities?"
1. The dragons may simply attack a settlement or village without warning, using their powerful abilities to wreak havoc and cause destruction.
2. The dragons may send an ultimatum or demand to the mortal races, stating that they must either submit to the dragons' rule or face the consequences.
3. The dragons may infiltrate the mortal races' organizations and try to sow discord and distrust, in an effort to weaken the mortal races' defenses before launching a full-scale attack.
4. The dragons may try to provoke the mortal races into attacking them first, perhaps by stealing valuable resources or committing other acts of aggression.
5. The dragons may work behind the scenes to destabilize the mortal races' society, perhaps by manipulating political leaders or creating economic chaos.
Ultimately, the specific way that the dragons initiate hostilities could depend on the dragons' motivations and goals, as well as the specific setting and circumstances of the campaign.
It was at this point that I realized it would take a considerable amount of chatting with the machine and teaching it the rudiments of a particular campaign world before it could go beyond generalities that would be obvious to any experienced GM, and I didn't have that kind of time. But maybe people on the boards do. Has anybody tried using it as a GM tool, perhaps to help with writer's block?
It won't write adventures for you, but it seems like it could handle intellectual grunt work, like spitballing ideas that the GM can then refine. I also tried to see if I could get it to write descriptive text for NPCs:
"If I were to describe the leader of the dragon armies in the style of a novel, how would I do it?"
"The leader of the dragon armies was a massive black dragon with piercing red eyes and wings so vast they seemed to blot out the sun. His scales were as hard as steel and his breath could melt steel. His name was feared across the land, and tales of his brutal conquests and ruthless tactics had spread far and wide. He was a being of pure power and ambition, driven by a burning desire to conquer and dominate all who opposed him. But beneath his fearsome exterior, there lay a deep intelligence and cunning that had helped him rise to the top of the dragon hierarchy. His enemies trembled at the thought of facing him in battle, but to his allies, he was a fearsome ally who would stop at nothing to achieve his goals."
Now, that description was a little middle-schoolish, and the AI didn't realize that black dragons breathe acid, not fire, so I went over it again making a few changes by hand to make it less clunky:
"The leader of the dragon armies was a massive red drake with piercing topaz eyes and wings so vast they seemed to blot out the sun. His scales were as hard as iron and his breath could melt steel. His name was feared across the land, and tales of his brutal conquests and ruthless tactics had spread far and wide. He was a being of all-consuming ambition, driven by a burning desire to conquer and dominate all who opposed him. But beneath his ferocious exterior lay a subtle intelligence: a cunning that had helped him rise to the top of the dragon hierarchy. His enemies trembled at the thought of facing him in battle, but to his allies, he was a fearsome comrade who would stop at nothing to achieve his goals."
Neither of them is winning any Nebula Awards, but when you're trying to juggle DM'ing with job/school/family, you're not trying to. You just want to prep your descriptive italicized text paragraphs with minimal fuss.
Anybody else tried using AI for something like this?
| 2 people marked this as a favorite. |
The problem with the AI is that it's scraping data from the internet, and you get a pretty bland average from the result. "An army of dragons attack everyone" is like... D&D movie bad. And not the new one.
If you're going to do this, I think you'll have better results if you include a bit of something specific and interesting in the prompt. Base the setting on ancient Sumeria or something.
... Buuut, if you're short on time and have twenty bucks (or four generous players with five bucks), I don't think anything is going to beat a published module written by real people.
|
| 2 people marked this as a favorite. |
Time for a hot take: AI is coming for everyone, absolutely EVERYONE who doesn't do physical manual labor that involves manipulating products to manufacture, refine, or deliver them and even in those cases, they too will see insane and continental shifts in how things work from the bottom up.
Folks have to remember that the whole AI Image generation stuff has only been openly public (and as a consequence getting trained on a larger segment of data and coached by users) for about three months now. ChatGPT only went public 18 days ago and in just this last week, after being played around with by D&D nerds for a few weeks it has almost perfected the art of generating 5e statsblocks and descriptions for creatures and NPCs. EIGHTEEN DAYS.
Folks in this and many other industries are absolutely SCRAMBLING and losing sleep over just how revolutionary AI improvements that have gone public are and the growth and power of these things is, well, calling it exponential is a MASSIVE understatement. There are some publishers pulling their hair out, image hosting/commission websites wringing their hands about what to do, and even programmers who are PANICKED about how fast the ENTIRETY of GitHub has been devoured to the point where amateur yet somewhat usable code is being created on demand by users who simply ask for what they want (so long as they know how to phrase the request).
Let me cycle back to something again because it bears repeating. this was launched and the more use it gets the better it does.
There are giants in the TTRPG industry as well as hundreds of others that post on social media on a more or less daily basis about the results they have, fear over what it all means, and in some cases very dire and seriously assertive opinion statements about how if we as a species do not crack down on, regulate, tag/identify, codify, and limit the use and growth of these tools that it will quite literally put 90%+ of the people in said industry out of a job overnight given that in less than a few minutes someone with the know-how for the tools (I'm talking TODAY) can whip out materials that would have taken a single person HURDREDS OF HOURS worth of labor in order to get something that is about 3/4s of the way toward being a very VERY solid "rough draft" for pretty much any type of writing, creative design, coding snippet/mod, bit of art, adventure, story synopsis, optimized travel/delivery/mail route, music, and just about anything else.
To directly answer your question, yes, I have tried using these tools and to SPECTACULAR effect. I'm no "the sky is falling" kind of doomer but what I've seen versus the effort I put into generating things for RPGs, Art, Music, Storytelling, and (this one REALLY GOT ME) even asking for ChatGPT to write a full and professional BUSINESS PLAN/PROPOSAL I'm completely astonished. People in EVERY industry are going to HAVE to pivot to allowing and adopting these as resources and tools that improve the way they work in terms of speed, tone, type, quality, size, scope, etc, and FAST or they are going to absolutely get demolished.
Legislation and AI identifying techniques are NEVER going to be able to catch up with the pace of advancement that is currently ongoing let alone the breakneck acceleration that it is getting from the influx of new users. There are certainly ethical questions about all of this that are quite troubling, esp from people who make a living off of producing words, art, and code for a living but to be totally frank about it, it's too late, the shoe has not only already dropped but it's miles behind us and there is no catching up. This is going to HAVE to be folded into the toolkit of every creative in every industry across the board and QUICKLY or they're going to be left in the dust as they spend dozens to hundreds of hours of work to put out something that is, perhaps, 10% "better" than the output that AI can give you in only a few minutes and in less than a year we will not only get to the point where AI generated materials are indistinguishable from the "real thing" but the AI generations will start to pull ahead such to the effect that they are of such HIGHER quality of what a human can do (by spending 100-100,000 x more time than the AI) at practically zero cost.
There are not going to be any laws that can be put in place or enforced to stop this, the momentum and force in this field is far too great, and the only hope left is for the creatives and companies that are threatened by it (nearly everyone) to latch onto it and ride the advancements as far as they can until their arms are ripped clean free of their sockets. Scary/exciting/weird times are ahead and RPGs are one of the areas that are almost certain to be impacted first. My only hope is that Paizo doesn't fall into the absolutely braindead trap of thinking they can stop this speeding bullet train by insisting their staff not use, rely on, adopt, or support it because like it or not, within the next few years all of their writers, artists, and other creative staff (secretly or not) are going to be leveraging these tools to help write, refine, and improve their work, all while spending less than half the time doing it than they normally would, either that or they decide to stick to their guns and refuse to deal with anyone who doesn't make "pure" contributions in order to get sub-par results that take 1000 - 100,000 times longer to create than they would get if they just embraced this new technological revolution that is coming to pass because of AI... and we aren't even to the point where we have real functional scale model quantum computing yet which is going to take the learning and speed of the system and catapault it into the next solar system in terms of raw power and speed.
| 1 person marked this as a favorite. |
So ChatGPT has been all over the news lately and I'm wondering if anybody has any experience using it as a Gamemastering aid.
Never heard of it before. I don't browse/listen to NPR very often though. I'd check it out but I'm not signing up for an OpenAI account.
Time for a hot take: AI is coming for everyone, absolutely EVERYONE who doesn't do physical manual labor that involves manipulating products to manufacture, refine, or deliver them and even in those cases, they too will see insane and continental shifts in how things work from the bottom up.
Don't worry, the robots are coming for the physical manual labor jobs and the AI will be telling them what to do... Skynet is watching. [shifty eyes]
| 1 person marked this as a favorite. |
I haven't heard of the thing before either - though I have heard about AI and machine learning, so the concept isn't new.
From what you have described, it basically it sounds like an automated plagiarism device. At least as far as scraping GitHub.
and even programmers who are PANICKED about how fast the ENTIRETY of GitHub has been devoured to the point where amateur yet somewhat usable code is being created on demand by users who simply ask for what they want (so long as they know how to phrase the request).
Aye, there's the rub.
When you have to make your request in C, creating an executable program as a result isn't difficult. That technology has been around longer than I have been alive.
Also, someone has to validate that the resulting program is actually suitable for the purpose it is intended for.
| 1 person marked this as a favorite. |
The problem with the AI is that it's scraping data from the internet, and you get a pretty bland average from the result. "An army of dragons attack everyone" is like... D&D movie bad. And not the new one.
If you're going to do this, I think you'll have better results if you include a bit of something specific and interesting in the prompt.
I know, if I was actually going to prep for a campaign I would do that. The above was literally just my first fifteen minutes with the thing.
| 1 person marked this as a favorite. |
My favorite thing I've seen so far was this article about having it stimulate unix. There's a level of versatility and power to this system that's only going to get crazier with time aside from funny little things like this.
Maybe I'll have my players chip in with me and I'll see if I can't feed it some foundry modules to speed up any custom work I'll need in the future.
| 1 person marked this as a favorite. |
I suggest that everyone who hasn't played with, used, researched, or interacted with at least the cool free open-source demo versions of this stuff take a bit of time to do so. Before I launch into it, I gotta switch my profile/avatar and hit you with a spoiler...
It took about 100 years for US households to go from 0% access to running water to get it near 100%. In 55 years household electricity in the USA went from about 1% to 99%. Home AC/Heating took about 40ish years to get to around 90%. The Radio took about 20 years to hit nearly 100%. The refrigerator similarly took about 20 years to do that same.
Now, compare this against the home computer that went from about 1% to 95%+ in 15 years. Smartphones STARTED at around 15% INSTANT adoption only about 10 years ago and are already approaching 90%.
This kind of AI is improving at a rate that absolutely DWARFS every single one of these by whole orders of magnitude in terms of how swiftly it is improving, and more to the point, it's is being done AUTOMATICALLY with little to no direct programming or intervention by human beings beyond ensuring that the computers and servers that host them are kept operational and powered. We are set to see this kind of technology go from 0% from a few years ago to the near 100% mark in what is most likely to be a timeframe of less than a decade. Let me reiterate that, the tools are almost certain to be something that everyone not only has access to but also actively does use on a regular if not daily basis by EVERYONE with access to a function and connected computer by around 2030. The stratospheric pace of self-taught advancement is ACTUALLY something unlike anything else humanity has ever seen before and this is by no means hyperbole. Add onto this the fact that is all FREE and has no manufacturing component to create whatsoever, something that was never the case for any other similar revolutionary tech developments we have seen in history with the possible exception of spoken language... which we have no documentation or knowledge of how it happened anyhow given it predated written word.
The memes of "the robots are coming to take your jobs" was partially true and made huge impacts on all sorts of industries but until recently the creatives were more or less safe as thinking, thought, expression, communication, writing, editing, and everything else somewhat related weren't something robots could easily or, more importantly, affordably do... but that is just not the case anymore and it will soon become widely and wildly apparent that biology and life, in general, are abominably awful at these things (at literally thinking) compared to refined, coached, and self-taught computer models. Things are not quite THERE yet but rest assured that there is absolutely ZERO CHANCE that global industry headed by capitalist corporations, nations, governments, and legislators are going to reject this kind of tech advancement as a means to be more efficient, assertively correct, faster, and all the while doing so without having to involve the costly oxygen wasting organic bugs in the code that are human beings.
As is, RIGHT NOW, with only the free tools one could spend an afternoon inputting creative directions, refactor what they're asking, prompt a few times, and have the systems refine/recreate the results until they get enough output that is near professional human quality to take and pass off to professionals to tidy and clean it up with a few passes and in doing so they would save dozens of emails, weeks of time, salary/contract costs, bureaucracy, spelling errors, personal bias, interpersonal conflict, human foibles, and at the end of the day they will end up with something that humans DID make with the help of the tools in a fraction of the time, energy, and money that it would otherwise cost. If I were a betting man, I'd wager that someone with enough experience knowing exactly how to submit queries could come up with a meaningful, solid, and substantially complete rough draft for an entire 800-page novel complete with initial draft illustrations for the cover/back and every single chapter therein in less than a day. This is the kind of thing that takes real and grizzled veteran authors MONTHS or DECADES to accomplish and it could be knocked out, almost certainly, in only a single day. Now, certainly it would then need to be digested by real humans (that is until we get more refined "editor" AI that can vet work like a real expert critical human would), read, submitted for revisions, and then hammered into a final shape but the first part being done automatically with mere simple directions and instructions all while the author sits at their desk munching down on some tostadas, typing with one hand, and binging three episodes of Black Mirror...
Now, onto the question... GMs... they'll probably be "safe" for a while as folks very much, I think, will prefer to actually play with real people but I'd bet in the next ten years or so that just about EVERY RPG system is going to be able to be taken by a player or group of them and requested to just be run by a computer or AI and it will simply just work. Authors, artists, programmers, middle managers, payroll folks, accountants, tax professionals, actuaries, and eventually even just about every legal profession is almost certain to be replaced or otherwise SUBSTANTIALLY supported by this tech. I'd even go so far as to say that there are probably going to be AI politicians or in the very least, humans that use AI to create their identity, platform, and politics because this whole thing is like discovering fire all over again. This time it isn't fire we discovered (a way to externally create heat instead of by way of biological digestion), it is THOUGHT and EXPRESSION that we discovered and it is going to be just as impactful on society, labor, and humanity as a whole as fire was.
| 2 people marked this as a favorite. |
As long as an AI is still running an algorithm under the hood, then it doesn't actually want - no matter how much it might look like it does. It doesn't care if it is successful or correct, it doesn't have desires, or wishes, and it doesn't innovate. Someone with real intelligence has to provide all of that first. The AI can be a cool and useful assistance tool, but not the primary driver of the project. If left to itself it will do absolutely nothing. That is simply the nature of algorithms.
There is also Rice's Theorem to deal with before an AI can start to improve itself Skynet style. In addition to actually wanting to improve itself in the first place.
| 1 person marked this as a favorite. |
That smells way too much like fearful optimism to pass the sniff test for me, just personally speaking, but only time will tell to prove any one theory about how it will advance correctly (or more likely nobody is able to predict exactly how it will all shake out).
Personally, I have no hangups revolving around the divinity or unique nature of intelligence, sapience, or creative spark that life as we know it grants certain earthen creatures so I don't foresee anything at all standing in the way between self-learning computation and what we view as sentience or intelligence, after all, what is already in circulation quite literally is rewriting and in some cases even rewiring itself to become more efficient and better at understanding hundreds of trillions of times faster than the required evolutionary timeline that life requires.
| 3 people marked this as a favorite. |
I have no interest in having a robot made by dudes who do too much coke for a living steal random reddit and stackexchange posts for me, thank you. If I want an adventure involving a dragon, I'll ask something that knows what a dragon is.
| 1 person marked this as a favorite. |
Here... let me just show you. I'm using the BAREBONES super minimum demo tool with no paid component, free instant signup, and access to practically ZERO improvements that have been made in the last two and a half weeks.
I generated this on the FIRST TRY and removed NOTHING. I gave it a simple 5 sentence description of what I wanted. A Paizo PF2 Adventure Path fighting titans that threaten to destroy the kingdom of Highhelm in which the party investigates raises an army against and puts to rest the threat to the kingdom with a target length of aprox 1800 words.
Table of Contents
I. Executive Summary
A. Summary of Project
B. Overview of Business Plan
II. Company Overview
A. Mission Statement
B. Vision Statement
C. Company Structure
III. Adventure Path Overview
A. Introduction
B. Plot Overview
C. Characters’ Motivations
D. Locations and NPCs
E. Development and Publication Timeline
IV. Market Analysis
A. Customers
B. Competition
C. Promotional Plans
V. Financial Plan
A. Start-up Costs
B. Assumptions and Projections
C. Capitalization Requirements
VI. Strategic Plan
A. Acquisition Strategy
B. Expansion and Growth Strategies
VII. Action Plan
A. Implementation Plan
B. Risk Assessment and Mitigation Strategies
VIII. Management Team
A. Leadership and Management Roles
B. Key Player Profiles
I. Executive Summary
A. Summary of Project
The proposed Pathfinder Second Edition 3-part Adventure Path series will feature a party of scrappy peace-loving peasants who must rise to the challenge of a cataclysm caused by seven ancient sleeping titans who wake beneath the surface of Golariion and in their waking state begin to stand, an action that topples an entire mountain range, destroys the Dwarven Capital city of Highhelm, and creates a massive inland mudslide that destroys dozens of underground cities. The party should be first level in volume one and attain level 4 by completion, start at four in the second part, and by the end of the third volume attain level 8 as they raise an army to put the drowsy giants back to rest, alive or dead. This series will use characters and settings from the Pathfinder Second Edition game world and will be written for a general audience as well as experienced players of the role-playing game. The target audience is adults aged 18 to 35 who are interested in fantasy role-playing games, and the goal is to give them a unique and exciting adventure series that can be used both as an introduction to the game and an engaging story arc for continuing players.
B. Overview of Business Plan
This business plan provides an overview of the proposed Pathfinder Second Edition 3-part Adventure Path series, including an introduction to the project, a market analysis, a financial plan, a strategic plan, an action plan, a management team, and appendices with additional information as required. The goal is to provide Paizo with the information necessary to make an informed decision about the viability of this project and its potential for success in the marketplace.
II. Company Overview
A. Mission Statement
Our mission is to create a unique and engaging Adventure Path series for the Pathfinder Second Edition game world that will inspire readers to explore their imaginations and engage with the game world in new ways. We want to create a story arc that will be accessible to newcomers to the game as well as experienced players, and provide a thrilling journey that will keep readers entertained for years to come.
B. Vision Statement
Our vision is to become the premier source for exciting and immersive adventures for the Pathfinder Second Edition game world. We will strive to create stories that capture the spirit of adventure and challenge our readers to think outside the box as they explore their imaginations.
C. Company Structure
The Pathfinder Second Edition 3-part Adventure Path series will be written and developed by a team of writers and game designers under the direction of an editor and creative director who will be responsible for overseeing the project from start to finish. This team will have dedicated resources for development, editing, illustration, layout, and marketing in order to ensure that the project is completed on time and on budget.
III. Adventure Path Overview
A. Introduction
The proposed Pathfinder Second Edition 3-part Adventure Path series will feature a party of scrappy peace-loving peasants who must rise to the challenge of a cataclysm caused by seven ancient sleeping titans who wake beneath the surface of Golariion and in their waking state begin to stand, an action that topples an entire mountain range, destroys the Dwarven Capital city of Highhelm, and creates a massive inland mudslide that destroys dozens of underground cities. The party should be first level in volume one and attain level 4 by completion, start at four in the second part, and by the end of the third volume attain level 8 as they raise an army to put the drowsy giants back to rest, alive or dead. This series will use characters and settings from the Pathfinder Second Edition game world and will be written for a general audience as well as experienced players of the role-playing game. The target audience is adults aged 18 to 35 who are interested in fantasy role-playing games, and the goal is to give them a unique and exciting adventure series that can be used both as an introduction to the game and an engaging story arc for continuing players.
B. Plot Overview
In Volume One, the party begins their journey as they investigate the mysterious events occurring around them as a result of the titans’ awakening. They must then travel across Golariion, discovering ancient artifacts and uncovering secrets about the titans and their power. Along the way, they will battle fierce adversaries and unravel generations-old mysteries as they try to find a way to put the titans back to sleep.
In Volume Two, the party must use their newfound knowledge to unravel the mystery of the titans’ origin, find a way to put them back to sleep, and protect Golariion from the devastating consequences of their awakening. They must also find a way to rally the people of Golariion behind them and build an army to stand against the titans.
In Volume Three, the party must lead the army against the titans and put them back to sleep, dead or alive. As they battle their way through hordes of monsters and face off against powerful villains, they must also discover the truth behind the titans’ mysterious power and how it can be used to protect Golariion from further destruction.
C. Characters’ Motivations
The party’s primary motivation is to save Golariion from the consequences of the titans’ awakening and find a way to put them back to sleep. Along their journey, they will also discover their own inner strength, uncover secrets about the titans, and ultimately protect Golariion from further destruction.
D. Locations and NPCs
The party will travel across Golariion, visiting a variety of locations such as cities, towns, dungeons, caves, ancient ruins, and more. They will encounter numerous NPCs such as merchants, adventurers, locals, and powerful figures. Each location will feature detailed descriptions of its inhabitants and encounters that will challenge the party’s skills and abilities as they battle their way to victory.
E. Development and Publication Timeline
The Pathfinder Second Edition 3-part Adventure Path series is scheduled to begin development in Spring 2021 and will be completed by Fall 2021. The books will then be published in Winter 2022.
IV. Market Analysis
A. Customers
The primary target audience for the Pathfinder Second Edition 3-part Adventure Path series is adults aged 18 to 35 who are interested in fantasy role-playing games, as well as experienced players of the game. These customers are likely to be familiar with the game world and will be looking for an exciting story arc that can be used both as an introduction to the game and an engaging story arc for experienced players.
B. Competition
The primary competition in the fantasy role-playing game market is other established game companies such as Games Workshop and Fantasy Flight Games. These companies have established fan bases and are well known for their adventure paths and other products related to role-playing games.
C. Promotional Plans
The promotional plan for the Pathfinder Second Edition 3-part Adventure Path series will include advertising in print media, social media campaigns, sponsored events at game conventions, and word-of-mouth promotion from existing customers. Additionally, we will partner with game stores and other retailers to promote our products through in-store displays, special offers, and discounts.
V. Financial Plan
A. Start-up Costs
The estimated start-up costs for the Pathfinder Second Edition 3-part Adventure Path series will include development and editorial costs, illustration and layout costs, printing costs, and promotional costs. These costs are estimated to total $60,000.
B. Assumptions and Projections
The sales projections for the Pathfinder Second Edition 3-part Adventure Path series are based on the assumption that each volume of the series will sell at least 10,000 copies at a retail price of $29.95 per volume. If these assumptions are met, the total sales could reach $899,500 over the course of three years.
C. Capitalization Requirements
The total capitalization required to fund this project is estimated at $60,000. This amount should cover all development, editorial, illustration, layout, printing, and promotional costs associated with the project. Additionally, an additional $10,000 is recommended for contingency funds in case of any unforeseen delays or expenses that may arise during the project.
VI. Strategic Plan
A. Acquisition Strategy
The acquisition strategy for the Pathfinder Second Edition 3-part Adventure Path series is to partner with existing game stores and other retailers as well as bookstores in order to ensure that our books are readily available for purchase in a wide variety of locations. Additionally, we plan to partner with online distributors such as Amazon in order to reach a wider audience and maximize our potential sales.
B. Expansion and Growth Strategies
Our expansion and growth strategies for the Pathfinder Second Edition 3-part Adventure Path series include creating additional books in the series as well as spin-off products such as sourcebooks, miniatures, and apparel. These additional products will help to increase our overall sales and extend our reach beyond our initial target audience of role-playing gamers.
VII. Action Plan
A. Implementation Plan
The implementation plan for the Pathfinder Second Edition 3-part Adventure Path series includes developing content, creating illustrations, designing layouts, printing books, and launching a promotional campaign. This plan is estimated to take six months from start to finish with a completion date of Winter 2022.
B. Risk Assessment and Mitigation Strategies
The risk assessment for this project includes potential delays in development, changes in consumer demand for role-playing games in general, competition from other game companies, and changes in market conditions that could impact our potential sales. We will mitigate these risks by staying abreast of industry trends, staying ahead of our competition with innovative products, and actively monitoring our sales numbers in order to adjust our marketing strategies as needed.
VIII. Management Team
A. Leadership and Management Roles
The management team for the Pathfinder Second Edition 3-part Adventure Path series will consist of an editor/creative director who will be responsible for overseeing all aspects of the project; a development team who will create content; an illustration team who will create artwork; a layout team who will design layouts; a printing team who will be responsible for printing books; and a marketing team who will create promotional campaigns.
B. Key Player Profiles
The key players on this project include:
• Editor/Creative Director – Responsible for overseeing all aspects of development, editorial, illustration, layout, printing, and marketing for the project.
• Development Team – Responsible for developing content for each volume of the series.
• Illustration Team – Responsible for creating illustrations for each volume of the series.
• Layout Team – Responsible for designing layouts for each volume of the series.
• Printing Team – Responsible for printing books for each volume of the series.
• Marketing Team – Responsible for creating promotional campaigns for each volume of the series.
Again, the above was generated for free, in less than a minute, and uses OLD DATA that isn't even hooked into systems that allow it to sample the rapidly increasing datasets such as user feedback about what is being generated, the quality of the rendering, or adaptations to the self-learning behaviors.
| 1 person marked this as a favorite. |
That smells way too much like fearful optimism to pass the sniff test for me, just personally speaking, but only time will tell to prove any one theory about how it will advance correctly (or more likely nobody is able to predict exactly how it will all shake out).
Don't take things too personally. I'm just trying to give my viewpoint since I do have a degree in the computer science field and a decade of practical experience.
I also find it amusing that their first featured example from that first link in this thread is basically a linter with a shiny skin. Linter programs have existed since 1978.
Don't get me wrong. It is a cool tool. It could eventually save a lot of long tedious hours of labor. And may indeed put people out of work who used to do those long tedious hours of labor.
I'm just trying to point out where the limits of computability are. There are things in computer science that have been proven to be actually impossible - not just very hard and out of our current technological power.
| 1 person marked this as a favorite. |
Hey, I'm not even what one might call an evangelist for the tech, that much is for sure, but I can guarantee that anyone crossing their arms and raising their nose at it is going to be in for one hell of a surprise when they realize that what we already have, in many ways, already surpasses the functional capabilities of even the most talented and knowledgeable humans alive and produces results in the timeframe of seconds compared to hours, days, months, or years.
If there is one technology that we shouldn't be underestimating it is AI, and by that, I specifically mean not only the use we could have for it but also the dangers and opportunities that will be opened by it.
Nothing you listed is beyond a half-decent DM sitting and pondering a situation for 5 minutes.
These are just filled in d8 plot generator tables that a lot of RPG corebooks ship in the Gamemastering chapter these days
I guess I'm not just floored by these kindergarten-level plot madlibs of "The (blank) are (blanked) to (blank) the (blank) by the (blank) of (blank) and have to fight the (blank)"
If they weren't generic, but were actual IP-related factions, AND the conflict flowed logically (e.g. - the Iomedaens clash with the Hellknights) then I might be impressed.
Yeah, that is absolutely the benefit of computers - doing the tedious work of solved problems faster than people can.
But it isn't going to be creating Pathfinder 3rd edition. It could do the formatting work (HTML, LaTeX, ...) - but couldn't be the creative director or decision maker. Also couldn't be the validator for traits, abilities, bonuses, and such - that would run afoul of the Incompleteness Theorems.
| 2 people marked this as a favorite. |
I haven't heard of the thing before either - though I have heard about AI and machine learning, so the concept isn't new.
From what you have described, it basically it sounds like an automated plagiarism device. At least as far as scraping GitHub.
Gotta say, "automated plagiarism device" is a great phrase that I will be using in the future to describe certain AI (especially the current issues in the art community with AI drawings).
-----
RE the AI Apocalypse:
There's two ways to take this.
The first is that the only jobs left in the dystopian future will be 1) Owners who make the profits; and 2) Programmers to refine the software as desired by the Owners. Everyone else will be in desperate poverty as they can contribute nothing of value and only a minimal amount of programmers are needed.
The second is to recognize that there are certain fundamental limitations inherent in AI that can't be solved by technological advancement alone, and are very likely to simply be impossible to execute. While the advancement of AI is almost certainly going to cause massive upheaval & change in the working world, there are still a number of limitations that make it undesirable as anything more than a tool for a large variety of purposes regardless of how much complexity it adds.
These are just filled in d8 plot generator tables that a lot of RPG corebooks ship in the Gamemastering chapter these days
I guess I'm not just floored by these kindergarten-level plot madlibs of "The (blank) are (blanked) to (blank) the (blank) by the (blank) of (blank) and have to fight the (blank)"
If they weren't generic, but were actual IP-related factions, AND the conflict flowed logically (e.g. - the Iomedaens clash with the Hellknights) then I might be impressed.
It might be able to do that if it had been trained on Pathfinder adventure paths specifically, but I don't think AP pdfs made up a large segment of its training data.
As a fun aside, I asked it "How do I book plane tickets to Sokovia?" and got this response:
It seems pretty banal, but remember, when Age of Ultron hit theaters there was no computer anywhere in the world capable of detecting the concept of fictional locations and that you cannot travel to them.
| 1 person marked this as a favorite. |
I kind of feel sorry for you all thinking this wont ever be more than cheap novelty trick.
| 2 people marked this as a favorite. |
I kind of feel sorry for you all thinking this wont ever be more than cheap novelty trick.
That's not what I am saying at all.
I am saying that it won't completely replace skilled labor.
I'm also pointing out that this new thing isn't really all that new. I studied AI and machine learning in college over a decade ago. It wasn't even all that new then. The syntactically correct English is a cute touch though. It is also an improvement over what I saw 10 years ago as far as detecting silly things that people would feed to it as input.
For my own job: Code generators have existed for longer than I have been a software engineer. Linter programs have existed for longer than I have been alive. I use such tools on a daily basis. But computers still can't create software.
That is all that I am saying.
I kind of feel sorry for you all thinking this wont ever be more than cheap novelty trick.That's not what I am saying at all.
I am saying that it won't completely replace skilled labor.
I'm also pointing out that this new thing isn't really all that new. I studied AI and machine learning in college over a decade ago. It wasn't even all that new then. The syntactically correct English is a cute touch though. It is also an improvement over what I saw 10 years ago as far as detecting silly things that people would feed to it as input.
For my own job: Code generators have existed for longer than I have been a software engineer. Linter programs have existed for longer than I have been alive. I use such tools on a daily basis. But computers still can't create software.
That is all that I am saying.
Ok. I apologize. You had some pretty well thought out observations.
I kind of feel sorry for you all thinking this wont ever be more than cheap novelty trick.
All my needs are met and all my hobbies are paid for, so I don't need or desire your sardonic sympathy.
A short list of things that I have, in the past 5 years, been on the "wrong side of history" on according to internet commenters:
Bitcoin.
NFT's.
VR headsets
The "Metaverse"
Smartphone gaming
etc.
We only talk about the ones that got it right, we leave the ones that didn't on the dust-heap of history and we do it really quick.
After 30 days exposure to AI art in Deviantart/Artstation, I can already spot an AI-generated piece of art a mile away and I'm not even an artist. I'm not impressed by them -- why would I be impressed by an AI text prompt when I actually write as a hobby?
It is a fascinating subject. I was a teenager when Garry Kasparov was defeated by Deep Blue. And computers have only gotten better since then.
It is interesting to me, at least, that computers that play chess don't win because they want to. They win because they are programmed to. And in fact, they can be programmed to not play as well - to varying levels of believability.
Bots doing reasonably well most of the time
Bots essentially programmed to lose
Did I accidentally write the exact opposite of what I intended? I'd like to see an AI bot make those goofy mistakes......
I guess if you're really struggling with basic ideas for games, this might be useful, but the hard part, especially when it comes to published work, is fleshing it out
I just come up with the first thing that comes to mind. That's why I don't think a robot would ever do a human being justice. I'll base an adventure off a fever dream I had involving a Jimmy Buffet song and thaumcraft. How the hell is an AI ever going to make those connections? I don't even know how I made those connections.
I'll base an adventure off a fever dream I had involving a Jimmy Buffet song and thaumcraft. How the hell is an AI ever going to make those connections? I don't even know how I made those connections.Do you really doubt AI could make random and weird connections? O_o That is a thing it definitely and constantly demonstrates even now. That AI art all is a kind of an electrical fever dream, for example.
You have a very different meaning of random than I do. I wouldn't describe the connection between two unrelated things that MadScientistWorking mentioned as 'random'.
Computers and automation have replaced jobs all over the place since decades, nothing new under the sun. Just the same questions rising again: How will we make the transition from a world built around work to a world where work will become scarce?
| 1 person marked this as a favorite. |
TBH I like AI tools not for replicating high concept components of a campaign, but smaller ones.
Being able to quick generate short backstories or personal details about random NPCs is good for fleshing out corners of the world I wouldn't normally focus on, and gives me a springboard if a character unexpectedly becomes more important than I originally anticipated they would be. Procedural NPC generation is hardly new, but the tools have gotten pretty nice for it.
Then I can turn around and even generate a quick portrait for them too using that description.
These are really handy tools for adding a little bit of extra flavor to a world.
I'll base an adventure off a fever dream I had involving a Jimmy Buffet song and thaumcraft. How the hell is an AI ever going to make those connections? I don't even know how I made those connections.
AI really tends to have more of the opposite problem right now. Everything a lot of modern chat bots write tends to have an element of surrealism to it.
What it has difficulty doing is keeping track of things and iterating on them long term. It tends to fall into loops, or lose track of which character is named what and where they are. They've been getting much better at this, but it's still an issue.
But spinning off an insane idea you can riff off of into a campaign is something they're rather decent at.
| 2 people marked this as a favorite. |
Computers and automation have replaced jobs all over the place since decades, nothing new under the sun. Just the same questions rising again: How will we make the transition from a world built around work to a world where work will become scarce?
Automation will not reduce the amount of labor in a capitalist society, it will just funnel that labor in to whatever the people that control that automation want. And the kind of people pushing for AI automation do not have workers' best interests at heart, generally.