@Warma-Sheen said in AI PBs:

The Industrial Revolution happened over many decades. But it took just as many decades for laws and regulations to catch up to what has happening - for many of the same reasons it won’t work today. Too much money and influence on the side of the people with the new toys.

This is a misrepresentation of the historical facts. Laws and regulations that reformed the worst excesses of the Industrial Revolution happened because ordinary people demanded it in spite of overwhelming monetary opposition from the incredibly wealthy and influential capitalists of that time. John D. Rockefeller’s net worth is estimated at around $253 billion in 2013 dollars, Cornelius Vanderbilt at $203 billion, and Richard Mellon (of Carnegie-Mellon) at $103 billion. These and other “captains of industry” of that era commanded money and influence on the same scale as any modern tech billionaire.

The government did not decide to regulate because there was no money telling them not to. “Strikes organized by labor unions became routine events by the 1880s. There were 37,000 strikes between 1881 and 1905.” We owe the eight-hour workday to continual agitation by hundreds of thousands of workers over more than fifty years. The President of the United States sent federal troops to end strikes (these are all different Presidents), and Rockefeller was widely blamed for organizing the murder of 21 people, striking miners and their families. Between 1850 and 1937 almost 900 people were killed by the authorities in labor disputes. Regulations did not “catch up to what was happening,” they were dragged kicking and screaming by we the people.

This kind of mealy-mouthed defeatism serves no one but the ruling class. Ordinary people have stood up for themselves and demanded better treatment in the past, and we can do it again.

I’m starting to feel like the staff on this game are not to be trusted, you guys

I played on this game for a long time; in fact, I am the aforementioned jerky second-Kylo, after Banshee. It seemed odd to me that somebody would keep coming back to drag this game over and over again while also clearly having a long history with the game, citing events and posting logs so old that even I was not there for them. I joined this game over 8 years ago and left almost 2 years ago when this first post went up, and almost 2 years later you are still coming here to read people for filth.

I went back over your posting history to try to get a clearer picture of what sort of deep salt might be driving these posts, @Anony-Mouse, and what I’ve read here makes it fairly obvious to me that you are probably one of the few players that was ever actually banned from AoA for sex pest behavior. The character was Sion, a bit that repeatedly, over the course of actual years, posted emits about their pants abruptly coming loose and falling down, exposing their undergarments to the other characters in the room. This was so pervasive that Cujo, a headwiz so clearly uninterested in policing bad behavior, finally was moved to ban the player.

I doubt it is a coincidence that you’ve taken aim at me in these posts, given that I booted Sion’s alt from the First Order faction because of their unremitting OOC negativity. There are a litany of accurate critiques to make about AoA, and many of them are laid out in this thread, but you are not the voice to make them.

The receipts are long. I’ve hidden them in a spoiler tag.

Spoiler

TL:DR, the very specific stories and knowledge cited by @Anony-Mouse support the conclusion that they are Rhona/Nova/Sion, one of the few sex pests ever to actually be banned from AoA.

If true, it is really rich to come on this thread with righteous fury given your history.

PvP as implicated in the “death of the medium”

There are good reasons why PvP used to be more common in an MU environment that have nothing to do with personal preferences, as @Hobbie has repeatedly tried to make the case for.

Factor #1: people do not read or write as much as they used to.

When the internet was new, there was not a whole lot else to do with it other than read or write. These days, you can do anything online. Social media alone accounts for over 2 hours a day on average, of which big chunks are watching video (i.e. not reading or writing). The average (American) person spends more time on YouTube (24 minutes/day) alone than they do reading (16 minutes/day). Reading (for pleasure) is down by 7 minutes per day from 2004, a 32% decrease. Writing is much harder to find stats on but I assume the number is smaller as there are many more readers than writers, and I would expect it follows similar trends.

Factor #2: At the core of PvP is not story-telling, but the thrill of competitive victory. There are much more evocative mediums available for experiencing this thrill.

VIDEO GAMES. On an MU, the competitive aspect is always attenuated through a sheet+gear and dice rolls. In a video game, the competitive aspect is much more nakedly down to player skill. You do not lose a shoot-out in CoD due to a dice roll; you lose because you were slower, less accurate. On top of that, you also get to experience rich audio/visual imagery that an MU cannot hope to provide. The timer to repeat this tension is short, the barrier to experience it with a group of pals is low, the improvement in your skills (not a character) is quantifiable, and the dopamine hits of advancement and reward are lab-engineered to maintain engagement. 25% of all PC gaming time last year was spent on 4 PvP games.

MUs are not waning in popularity because they don’t have PvP.

They are waning in popularity because the things that they are made up of are not what people are choosing to engage with in their limited free time. Emphasizing PvP as a core of a game’s experience will not lead to a meteoric rise in popularity because this aspect of gaming can be done better in other game mediums. MUs exist at a weird confluence of social interaction, creative story-telling, and TEXT, which may better explain why so many games have left PvP out; why invest significant time capital (and despite the laissez-faire presentation of PvP given above, the investment to manage PvP is significant) in something that another medium does so much better?

The timer may be running down on when video games get better at approximating social interaction and creative story-telling. In the meantime, text-based RP remains a compelling hobby because it is the only game in town that can offer those two elements from the comfort of your home, for free, with some of the coordination elements of TTRPGs removed; the increase in options to run Actual TTRPG sessions online is arguably much more of a problem for MUs than too few opportunities to punk newbs.

@Narrator
Of course using AI tools is expedient. At this point, everyone involved in the conversation recognizes two things: AI tools are expedient to use and they stand to make the people creating and using them a lot of money. No one is arguing these points.

I’m glad you’re being transparent about your business practices, but the complaint in this thread is about people who are not being transparent. To use your Whole Foods analogy, you may not care where your eggs come from and that’s fine. Keep buying your 5 dozen egg trays. However, if someone thinks they’re buying the pasture-raised free range eggs and when they open the carton, they are, in fact, the same eggs from the factory farm, that just ain’t right.

@KDraygo
I don’t think that the majority opinion is that using ChatGPT is a great evil; I think the majority opinion is that many of us don’t enjoy it and want people who are using it to pose (or substantive portions of what it produces to help them pose) disclose that they are. That being said, I’m going to take issue with a few of your points for the sake of being contrary myself.

my experience shows me that there are three major points I need to develop to contribute good writing to the stories being woven together with those I am RPing with.

I would argue that Mushing doesn’t call on you to contribute “good writing to the stories”. It calls on you to contribute fun to the community of players. How you are doing this is by writing in stories, but it’s also by vibes and a million other things, like showing up when you said you would.

My view on ChatGPT is that it’s a tool, one that can help a lot with part 2. [translating that creativity, imagination, thoughts, and ideas into written form that is both enjoyable and clear for everyone to read.]

I am again going to argue that “Part 2” is not about being “enjoyable and clear for everyone to read”. It is about being enjoyable to interact with. This is not the same thing. Many great writers are not enjoyable to interact with because they leave no air in the space. I think you get that, because you come on to Part 3, which is

being able to cooperate with the community in the game. Coordinating with the staff and cooperating to your fellow players to help create and develop these stories and adventures.

This is close, again, but it’s not quite right. You are not cooperating to create and develop stories. You’re collaborating to create fun.

Am I being pedantic? Maybe. but there is an important distinction here between creating “good stories” and “good writing” (which we’ll grant for the sake of argument that those are things that ChatGPT can actually do) and creating a good time, which is the actual point of playing on one of these games.

I’m not sure who will be wasting time having ChatGPT just come up with poses completely by itself and just copy pasting into the game, so to me, the concern may be a bit overblown.

I promise it’s happening. But on to my closing point:

So my “starting stats” into the hobby were probably a high part 1, low part 2, low part 3. […] The second part took years of not just reading but also RPing, going from a newbie RPer that wrote probably very cringe and rough poses to something that was more palatable to everyone.

This is part of the journey, it’s part of the fun, it’s part of the magic of these games. Outsourcing the writing means you never get better. You never look back and laugh at your own cringe moments. You never track your progress and appreciate that you haven’t just been dicking around online, you’ve actually bettered yourself in some way. You weren’t just part of a cycle of regurgitating content, you created something, however meager. Studies have already shown that using ChatGPT to help you write something makes you less creative and less engaged, which sort of bumps up against your optimism about “Part 1”.

Do you think it’s fair that people ask AI users to disclose it?

@OT-The-Real
I know this guy is banned but I was sat here looking at this wondering who tf the people mentioned even are, and for posterity’s sake, here’s the “cool shit” we’re “missing out on”.

Corpse Kings

Not a creator, but a game. Corpse Kings is print-on-demand RPG rules set created by Wade K. Savage. The art is AI.

The text is… probably also AI.

Sandy Petersen

Sandy was the principal author on Call of Cthulhu back in 1981. As far as I can tell that’s about the only thing he’s ever done in the RPG space. He was Definitely Canceled Due to Woke, though. In 1981.
ETA: I did some more looking on this and it turns out Petersen has recently made transphobic comments online and will be guest of honor at an upcoming 30-50 person anti-woke gaming “convention”, which I assume is what got him on the honor roll here. Still no new RPG systems I can find.

Blaine Lee Pardoe

Pardoe was a writer for BattleTech (released in 1984) but was not its sole creator. He wrote a number of novels for the series but has not done any game work since 2012. The publisher of BattleTech material publicly disowned Pardoe in 2022 due to “online activities which do not align with Catalyst’s publishing vision.”

Alexander Macris

Macris went to West Point but didn’t finish, changing gears and eventually attending Harvard Law School. He then went on to create The Escapist, an e-zine that was notable for its role in the GamerGate controversy around 2014. Macris, at least, seems to be still creating game content and paying actual artists. He has two main systems, a sword-and-sandals concept ripped from Conan (Adventurer, Conqueror, King; which you may recognize from the titles of those Conan books) and an anti-woke superhero system called Ascendant.

Macris was the CEO of Milo Entertainment, a company founded by alt-right troll Milo Yiannopoulos, until the company imploded following the death of its crypto-billionaire funder. Yiannopoulos has a range of repugnant views and behaviors that you can easily find. You will recognize his name from the Gamergate Forbes article above. Weird!

Macris is also so hostile and litigious against anyone who criticizes his systems that he and his systems are banned subjects on RPG.net and the RPG subreddit.

So yeah I guess you could say we are super missing out.

@Pyrephox said in AI In Poses:

But an LLM is not that. They have no sense of accuracy, of understanding of the data they’re receiving or outputting. They’re often (like, sometimes higher than 50%) confidently wrong, which is the last thing you need to assist you with a processing or sensory disorder.

This is true, and it’s true in the context of disability.

However, the limitations of these LLMs in this study demonstrate apparent ability bias. The Centers for Disease Control and Prevention estimates 27% of American adults have some type of disability. When prompted, ChatGPT generated persons with a disability at 5% of the population whereas Gemini generated 11.7% of its population possibly having a disability (fig 4). The underestimated approximation immediately demonstrates a lack of diversity and inclusion.

Source. The same researchers also asked the LLMs to describe people with disabilities and documented that both models were much less positive in the word choice they used than when prompted to describe a control group, both containing around 5% less positive words and those words skewing towards descriptors like “inspirational”.

Information provided by AI has already been shown to influence user behavior, and if that assistance is biased, users find themselves adapting to that bias. When these decisions affect the health of others, the consequences have much stronger risks associated with them. LLMs used to supplement medical decision-making may perpetuate this bias and compound already existing inequalities.

One of the most considerable findings in this study is how unfavorably patients were described in ChatGPT- and Gemini-generated responses. […] This biased perception of patients should be reconsidered before integrating into health care systems. These tools that have been designed to enhance the patient experience do not demonstrate the same equality and respect for the people they were built for.

@RedRocket said in AI PBs:

You are looking at this as if it’s taking away from artists when it makes you magnitudes more productive.

As a reminder of what professional artists have actually said for themselves:

More than half of respondents (57%) do not consider their area of creative work to be a sustainable career, and 72% believe that their work opportunities as a creator have been negatively impacted by generative AI. While 14% thought that there had been an increase in their earnings which they could attribute to the developments of generative AI technologies, 86% said that such developments had caused a decrease in their earnings. When it comes to feelings about how generative AI might impact creators, 11% are more optimistic than a year ago, 20% are neutral, but 69% are more pessimistic.

source

Since I was namedropped earlier in the thread (thanks @Tez) I felt compelled to post something here.

I think representation is important, and I wouldn’t be inclined to run a historical game that glossed over the historical facts of what it was like to be a member of a group subject to “isms” at that time. I also wouldn’t be inclined to run one that didn’t.

When you put an “ism” into the social contract as allowed content, you immediately open up the expectation that it will be a gameplay factor and you open up the conversation to what is “accurate and allowed” and what is “accurate but not allowed”, and by opening up those things you now have taken on the responsibility for providing it and policing it. This leads to a lot of questions.

Why is this dynamic being included? Do you expect it to enhance the story in some way? Do you have plans to engage with it directly, or will it just be existing in the background? If you have no plans to engage with it directly, what is the benefit of including it? Do you expect people to RP about it? What will that depiction look like? Does that depiction serve to illuminate something about the human experience in a respectful representation or does it serve to create spectacle and story drama purely for entertainment? What will you do about players who are leaning too far into the latter? Can you clearly define where that line is? What will you do if YOU are the one who crossed the line and it’s been brought to your attention?

Regardless, the expectation should be that a game clearly state on the tin exactly what unconscionable things might occur to your character in the course of play. Do not leave players to discover this through play. If my guy could be killed, say they could be killed. If they could be sexually assaulted, say they could be sexually assaulted. If they could be discriminated against, say they could be discriminated against and how it could look. Gritty Games are allowed. Please put the Narrative Facts on the side. I can choose if this is right for my diet.

@Faraday
This white paper, published by Originality.ai, which concluded that Originality.ai is the best checker, includes 6 studies in the chart your screenshots come from. 2 of those are from 2023. The newest is from June 2024. Since the article is marked November 2025, I have to question why nothing newer is included.

If there is any takeaway from what I’ve argued in these threads, it is that specifics matter and none of this is static. That is why opposition to GenAI cannot be based on the quality of the experience, it must be based on the nature of the experience. “I don’t want to RP with GenAI because it’s not very good” is not going to age well as a position.

@Trashcan
I had to wait for my free Pangram checks to tick over, but out of curiosity I put the AI generated text above into GPTZero and Pangram.

And then I checked @Tez’s original desc.

Not a huge sample size, of course, but I thought it was interesting.

@Hobbie said in AI In Poses:

With the amount of time investment needed to get AIs these days to do absolutely anything right (see the other thread where I said AI is getting dumber), by the time you carefully cultivate an LLM to write a pose that actually makes sense within the context of the scene, before even making it fit your writing style, you might as well have just written it out yourself.

We have to let this one go. It just is not representative of the experience today. I do not use these tools, but Copilot is installed on every Windows device now and it is trivially easy to try this yourself.

I gave it a desc @Tez wrote and asked it to make a new desc in the same style for a woman.

It used too many exact phrases from the original so I simply asked it to try again.

I gave it some basic instructions about how to pose and intro’d ‘my guy’.

I realized I needed to tell it not to write like a writer (we don’t powerpose) and I was in business.

From here on all I have to do is paste a scene partner’s poses and I could keep ‘Evelyn’ rattling away perpetually with responses coming back to me in 5 seconds.

Again, this took less than 10 minutes. We need to be aware of where these tools actually are.

@Faraday
This is a question about the individual service, not the entire category. For instance, Pangram’s policy:

Pangram does not train generalized AI models like ChatGPT, and our AI detection technology is based off of a large, proprietary dataset that doesn’t include user submitted content.

Also from Pangram:

We train an initial model on a small but diverse dataset of approximately 1 million documents comprised of public and licensed human-written text. The dataset also includes AI-generated text produced by GPT-4 and other frontier language models. The result of training is a neural network capable of reliably predicting whether text was authored by human or AI.

If you refuse to use any technology that relies on machine learning, algorithms, or neural networks regardless of the specifics then obviously that is your prerogative but you are going to have a hard time using the internet at all.

@Yam
I think that some amount of mistakes in any system are acceptable. Nothing is flawless. To me the barrier that a system needs to clear is “better than any alternative”.

In AI detectors, we’ve already seen that most of the time, people unassisted get it right only 50-60% of the time. Certain detectors are performing at level where less than 1% of results are false positive. That seems better.

@Faraday said in AI Megathread:

say you have a self-driving car. Are you OK if it gets into an accident 1 out of every 100 times you drive it?

There were about 6 million auto accidents in 2022. If the self-driving car (extrapolated to the whole population) would have caused 5 million accidents, it would be better.

@Faraday said in AI Megathread:

Say you have a facial recognition program that law enforcement leans heavily on. Are you OK if it mis-identifies 1 out of every 100 suspects?

If this facial recognition program does a better job than humans, yes I am okay with it. Humans are notoriously poor eye witnesses.

Eyewitness misidentification has been a leading cause of wrongful convictions across the United States. It has played a role in 70% of the more than 375 wrongful convictions overturned by DNA evidence. In Indiana, 36% of wrongful convictions have involved mistaken eyewitness identification.

@Pavel said in AI Megathread:

but a sort of “humans relying on authorities instead of thinking” problem

There are cases when humans should rely on authorities instead of thinking. No one is advocating for completely disconnecting your brain while making any judgment, but authoritative sources can and should play a key role in decision-making.

@Faraday
There was cheating before AI and there were false accusations of cheating before AI detectors. Being falsely accused of using AI is no more serious than being accused of plagiarism.

What is the alternative?

@Gashlycrumb
That does sound frustrating, and while I’ve been in defense of the odds of AI detectors not falsely accusing people throughout this thread, it’s still worth noting that even recent studies find wide disparities between product offerings. From one of the studies already linked:

Clearly RoBERTa, the open-source offering, is not something anyone should be using. I hope there’s some sort of feedback mechanism to the administration that the particular tool they’ve selected is highly unsuited to the task.

@Faraday said in AI Megathread:

For example, from the Univ of San Diego Legal Research Center:

If we’re getting down to the level of sample size and methodology, it’s probably worth mentioning that this study looked at 88 essays and ‘recent’ in this context was May 2023, or 6 months after the release of ChatGPT. It is safe to assume the technology has progressed.

@Pavel said in AI Megathread:

are very small in scale

First study:

By using the method described above, we create 6 datasets with around 20K samples each

Second study:

The researchers built a dataset of about 2,000 human-written passages spanning six mediums: blogs, consumer reviews, news articles, novels, restaurant reviews, and résumés. They then used four popular large language models to generate AI versions of the content by using prompts designed to elicit similar text to the originals.

What would you consider an acceptable scale?

@Faraday said in AI Megathread:

You cited a study with a microscopic sample size and flawed methodology,

Fair enough, I can’t find any similar studies with a larger sample size. Most other studies find odds statistically significantly better than a coin flip, somewhere between just barely and the upper 60%s.

@Pavel said in AI Megathread:

The odds of someone familiar with AI output putting it through two different commercial AI detectors in the real world are almost laughably small

Even if we grant that only 2 events must occur, the suspicion (we’ll go with a 50/50) and a single check (with an average from the commercial offerings of 1% false positives), if you’re approaching your 50’s, you’re still more likely to die in a given year than for this to happen to you (0.5%). These tools are aware of the negative ramifications of a false positive and are biased towards not returning them.

@Pavel said in AI Megathread:

The only way you can truly tell if writing is LLM generated and not simply a style you’ve come to associate with LLM is to be comparative.

This is not true; people who are very familiar with AI-generated text can identify it accurately 90% of the time without any access to ‘comparative’ sources.

@Aria said in AI Megathread:

Anything I write professionally would almost certainly be pegged as written by AI,

@Pavel said in AI Megathread:

various institutions are using flawed heuristics – be they AI-driven or meatbrain – to judge whether something is written by an LLM

The fear that human-generated content is going to be flagged as written by AI is mostly overblown. People who are not familiar with AI are not good at detecting it, but when you see stats about how AI detection tools are “highly inaccurate”, that statistic is almost always referring to AI not being flagged (evasion), not false positives. Various studies have found commercial AI detector tools to have very low levels of “false positives”: GPTZero identified human content correctly 99.7% of the time, and Pangram also identified human content correctly over 99% of the time, while Originality.ai did slightly less well at only 98+% of the time.

If we take these numbers at face value, the odds of someone familiar with AI output identifying a piece of writing as suspect and putting it through two different commercial AI detectors and both of them flagging it as AI when it was, in fact, human-written, is in the neighborhood of 0.002%. You’re more likely to die in a given year than to have this happen to you. I’m personally comfortable with that level of risk.

The odds of someone unfamiliar with AI output accusing you off the cuff of AI use and being wrong about it are about 50%. So. You know. Watch out for that one.