WotC Would you buy WotC products produced or enhanced with AI?

[Faolyn]

Yes, and all of those are technology assisting the keyboard player in doing things he-she can't otherwise do.

So by that logic, even a normal, non-electric musical instrument is assistive technology. Heck, even a pencil is assistive technology.

Congratulations, you have made a word completely meaningless.

No, that's not what assistive technology actually is. Here's the definition according to WHO:

Assistive technology is an umbrella term for assistive products and their related systems and services.

Assistive products help maintain or improve an individual’s functioning related to cognition, communication, hearing, mobility, self-care and vision, thus enabling their health, well-being, inclusion and participation."

[...]

Assistive technology is most needed by:

• older people
• children and adults with disabilities
• people with long term health conditions such as diabetes, stroke and dementia.

Musical instruments and autotuners aren't AT. Neither is AI.

 

[Jfdlsjfd]

AI cannot get the pose right or the colors that I wish even when I am descriptive.

You might be right about colors (as they can get very specific and AI training is usually done by specifying a blue coat, not a very specific shade of blue, at least so far), but poses have been solved quite a long time ago. Especially simple ones. You might want to use more powerful tools than just prompting for that, though.

 

[Faolyn]

Talent is a myth. What looks like talent is mostly the result of hard work and practice. Some people take to some endeavors faster or slower than others, but nobody is inherently good at anything.

Err... not so much. Some people do start off with a higher skill level in something. For instance, most people start drawing tadpole people at the age of 3-5. I started drawing them at the age of 18 months. I had a talent for art.

However--and this is the important part--without continued hard work and practice, my artistic talent would have gone nowhere. As it is, because I've gone long periods of time without drawing (thanks, chronic depression and ADHD), I'm a decent artist but not a fantastic one. Someone who started off with no talent but was much more diligent about practicing than I was would be a better artist.

 

[Jaiken]

but poses have been solved quite a long time ago. Especially simple ones. You might want to use more powerful tools than just prompting for that, though.

There may exist AI programs that can make poses that are specific, but they are usually behind a paywall that and it is not yet (up to my standard at least) able to make dynamic poses that I look for.

 

[Jfdlsjfd]

There may exist AI programs that can make poses that are specific, but they are usually behind a paywall that and it is not yet (up to my standard at least) able to make dynamic poses that I look for.

You're indeed right when it comes to online models. I think for getting the best results, tools (like controlnets) are more effective for models that you can run at home from your computer (which requires some time investment). Usually online generators (including paid ones) offer few options above just prompting a few words. They emphasize ease of use over getting exactly one has in mind (I'd say that many users don't have something very specific in mind, just general).

 

[Faolyn]

I think some of the disconnect is that you are viewing art as the primary creative endeavor, when for a lot of people it is an add-on. Many rpg users are primarily expressing themselves creatively through the text, through the world they create, etc., not through the art they include. They probably don't have any art at all, or art that doesn't fit their ideas or their setting. In these cases, AI is assisting people in their creative endeavors.

If I didn't use the term "art and writing" in that particular example, I definitely used it elsewhere.

A good writer can paint a picture using words. It's why most novels aren't illustrated.

Regarding both this and the previous statement, I think we need to apply the 'procgen map litmus test'. Having a program output a medium quality map that fits my ideas is helping my creativity, not preventing it. And it is ok that I'm not getting better at mapmaking, because the goal of my activity is not to become a good mapmaker.

If it "fits your ideas," it's not helping your creativity, it's doing it for you.

Not everyone celebrates inventions that save tedious labor. See, Luddites.

You mean the people who protested because their work was being done by people with lower skill levels and who were paid less, because those people could operate machines, but produced poorer work. Their actual complaints were against the employers, who hired those less-skilled/unskilled people so they could pay them less. Wrecking the machines were a convenient way to show their anger against the employers. If the employers had kept on the original employees and had continued to give them their original wages, the Luddites wouldn't have formed or protested.

Yes, that's exactly the case here.

I'd rather pay more for better quality produced by someone with skills than pay less for someone to write a prompt and click a button.

 

[The Firebird]

If I didn't use the term "art and writing" in that particular example, I definitely used it elsewhere.

A good writer can paint a picture using words. It's why most novels aren't illustrated.


If it "fits your ideas," it's not helping your creativity, it's doing it for you.

Ok Faolyn, I don't think we are making any more progress. If it isn't a good response to the map example I don't think I am going to find it convincing. Thanks for the conversation, despite the disagreement

 

[Charlaquin]

Thanks! Not my point, but good info.

Still, Newton's Laws . . . according to others in this thread . . . are still valid in the correct contexts. And either way, they are not "lies" but simply older scientific ideas that have been superseded by new knowledge.

The fact that a lot of folks, including myself, are not aware of this specific change . . . it's either not important if Newton's Laws still apply in everyday contexts that most folks exist within . . . or it is important and we need to improve our science education in schools and our science reporting in the media.

Yeah, calling Newton’s laws “lies” is certainly not accurate, though I think from context it was pretty clear that the term was being used for hyperbolic emphasis. The reality is that Newton’s laws are close enough to accurate that you can use them in most contexts that most people will encounter in their daily lives without issue. In that sense, using them is a bit like using 3.14 as an approximation of Pi - It works for most intents and purposes, but it is an approximation, and when you get into more advanced problems, you need to start using a closer approximation, like general relativity (which is also just an approximation).

The thing is, and I think this is part of the point @EzekielRaiden was trying to make, is that that’s true of all scientific laws. Any good science communicator will tell you, all models are wrong, but some models are useful. General relativity accounts for a lot of cases that the Newtonian model of physics does not. But we know it’s not the whole truth either, because it doesn’t work in the context of the very small, where you have to use quantum physics instead. And, likewise, we know quantum physics isn’t the whole truth, because we have so far been unsuccessful in using quantum physics to model gravity.

Part of beauty of the scientific method is supposed to be that it recognizes its own limitations. We can use it to find ways of modeling observed reality, which is very useful for many purposes. But we do ourselves a disservice if we forget that a model is not actual reality. It’s always just our closest approximation, until we develop a new model that explains more with similar complexity, or explains the same amount with less complexity (or explains more with less complexity, but that’s very rare). At which point the new model becomes our closest approximation. Like infinity, you can only approach truth, you can’t actually “count” to it.

 

[Charlaquin]

Err... not so much. Some people do start off with a higher skill level in something. For instance, most people start drawing tadpole people at the age of 3-5. I started drawing them at the age of 18 months. I had a talent for art.

However--and this is the important part--without continued hard work and practice, my artistic talent would have gone nowhere. As it is, because I've gone long periods of time without drawing (thanks, chronic depression and ADHD), I'm a decent artist but not a fantastic one. Someone who started off with no talent but was much more diligent about practicing than I was would be a better artist.

I think we agree, I’m just using more provocative language to express the point. Yes, some people are faster or slower learners at some things. But there’s no such thing as someone who’s great at something without continued effort. They might be ahead of the curve, but they still need to put in work to stay ahead of the curve.

 

[Clint_L]

And, theoretically there’s no reason you couldn’t learn to be as good as they are. It would likely take a great deal of time and dedication, which you may not have, or may not want to spend on such a pursuit. But in theory, it is entirely possible.

This is just not true. I have read through many, many psych-ed reports in my decades as a teacher, as well as doing a required minor in psychology with a focus on adolescent development. It is just not true that brains are all an equal tabula rasa with identical capacities. And the assumption that it is true has done a lot of damage to a lot of people. I know this is not your intent, but the notion that "all men are created equal" should not be interpreted to mean that all humans are the same.

People have different capacities. That is not debatable, that is a fact about human brains. And bodies. As a teacher, it is my job to help people maximize their capacity, and an important part of that (and the whole point of a psych-ed assessment) is helping them identify strengths that they can strive to maximize, and weaknesses that they can strive to minimize.

A person very close to me tests in the 98th percentile for verbal learning, but in the bottom 5 percentile for spatial/abstract reasoning, particularly in fields like mathematical theory. They are never going to be a mathematician, no matter how much work they put in, and it would be a miserable waste of their time and potential to try to force it.

In fact, this is what standardized education has done for generations of trying to jam all kind of learning "shapes" into a few simple boxes, and it has done immeasurable harm to generations of learners by convincing millions of brilliant people that they are stupid or can't learn, because their strengths were not valued at school, and their weaknesses were the benchmark.

This is not theory, or metaphysics. There are miles of research on learning and cognition. Each person is unique, and should be treated as such. That's the only way that they will thrive, and our society will reap the benefits.

"Time and dedication" are invaluable. But they are not all that matter. Every single person has unique capacities.

Edit: also, think about the implications of the statement that "theoretically there’s no reason you couldn’t learn to be as good as they are." In effect, this is saying that if I'm not Einstein, it's because I'm not trying hard enough. And it is this attitude that has hurt so many students for so long. I have plenty of students who work their butt off for a "C" on their Theory of Knowledge essay, and others who earn an "A" with a fraction of that effort.