Home Creators Posts Import Register Favorites Logout

Content

Hello everyone!

I know this is a controversial topic, and I would like to put some points for discussion (wood in the fire?).

To be clear, my point of view was formed by my experience and formation, as well as some other comments that I've seen. And to be real, it's not set in stone, and I might change it sometime. Also, what I think is not the absolute truth of what this is and how it's going to shape the world.

Why I'm making this topic?

A few weeks ago, Lychee announced its AI platform. I'm on their Discord server and have seen how Lychee employees were treated. Even though you disagree with AI, it does not give you the right to say certain things to Lychee employees, they are "nobodies" in a sense of what the company does with it's money. Harassing them will take you nowhere.

Respect is necessary so we can all reach a better place.

You dislike it; there are a lot of ways for you to make them notice your disapproval (as mentioned in the Tableflip Foundry community).

Am I going to drop Lychee?

As mentioned before, I'm on neutral ground when it's about AI models, and Lychee is currently the best slicer in the market. I also need Lys files to upload models to Lychee Library in the future. I don't pay them for sub.

Lychee Library is a complement to my current funds (far from Patreon or MMF, but every penny helps).

Why I'm neutral regarding AI

For starters, I would like to point out a few things that I saw that help me made my mind.

Today I watched a video from Artifex Maho, he was talking about 3d printed miniatures with color and how it would affect him and the hobby. The miniature itself does not affect him because he likes to paint, but he recognizes that some people don't like to paint miniatures but want them painted.

AI cost it's free! Yeah, about that. I've seen a video about a guy speaking about how AI it's working and how it will work in the future. AI is currently free because they are using it to create a dedicated userbase and fine-tune the tools; in the future, it will be charged (Lychee AI is a good example; you pay to create models)

The "AI Artist". In this video, a guy, shows a book that he made and opens at a art, and he asked if he was the artist because he gave the prompt, commissioned, and added to the book. In short, no, he could never be called an artist, and the same goes for the AI.

My formation as an Engineer does not allow me to think that all the jobs will be gone. Why? When we look at the industry, when we had manual labor, and replaced people with machines, the jobs were not gone; they shifted to something else. Where? Most modern machines require an operator or a team of operators to work, they also require a maintenance team. The machines were designed and built by a team somewhere, and spare parts and consumables will also need to be produced.

I personally don't believe in Equity, but I think this is the most possible way to see it as Equity

This is a generated image by AI. It should bring every character to the same level, but I guess AI considers it self taller than the rest lol. Anyway, for the sake of the argument, consider everyone at the same level because I will be talking about the result as a finished miniature, and the size of the character is related to a person's skill (I tried to make a few runs, but it did not work out, and I did not want to lose much time with this)

Classic, these guys are badass! I'm nowhere near their might. Sculpting things by hand and some master pieces with tools, putty, and their hands. A shout-out to Michelangelo and others who made some amazing sculptures with primitive tools and stones.

Software, we use 3d software and a 3d printer to achieve the same result as the Classic dudes. In this category, there are many levels of artists, and we might not always reach the same result, hence the Equity thing mentioned before.

AI, here, people who don't know how to sculpt ask for the tool to create something and have a miniature in their hands. The problem here is that, as you can see, AI can create "stuff," but it can sometimes produce some slop. Ark Aquatica trailer is a nice example of slop.

The problem here is that people call themselves AI artists, produce slop, and are trying to sell it. This is a scam (I took notice of this by a miniature painter, Pintando Miniaturas, he made a video about it). This indeed it's damaging to the consumer and also damages the reputation of sculptors who take their work seriously and check every bit of the miniature before shipping (but even still sometimes a thing or two passes unnoticed).

AI is getting good, but it's not at the point where you can close your eyes and produce your miniature without adjustment. Sure, it can create some simple things.

Another thing that I would like to point-out is that people that dig into AI 3d models probably would not buy a miniature or subscribe to a sculptor. I hope that with AI we might see a bit less pirates in the market.

AI in our lives

Most of us are using AI and don't even notice that we do. When we search the browser, it has an AI built in now. I'm using AI assistant to correct my English (I make many typing mistakes, English is not my main language).

I also consider Auto Support as AI, think about it, the software analyzes the model and places supports where it thinks it will print great (most of the time it will fail, that's why I don't use)

I give some ideas to Chatgpt and ask it to refine and suggest variations. In the end, I adjust everything, giving my touch. I'm lonely and have no coworkers (this looks sad lol, I wanted to say that I work alone), so I sometimes talk to ChatGPT as a colleague.

A few Ideas of how I think AI could be used (a tool for the hobby)

Kitbash, you present the AI 2 3d models and ask for it to mix some parts, maybe exchange some armors, weld vertices, and create a solid or hollow model. Swap weapons and etc.

Reposing. Sometimes you have a 3d model, but you want it in another pose. I think it would be cool if you could load a model in the software and adjust its pose like Heroforge to suit your needs.

I don't think these tools exist, but I think it would be awesome for a user to have. I don't mind you kitbashing my models, and I want to see some epic stuff painted that I helped to create (even if my contribution is a small toe).

From my view, the Problem of AI

The problem of AI is not the AI; for me, it's the lack of people adjusting and correcting the AI. If you watch Ark Aquatica AI slop trailer, you will know. For that trailer to be published, people forgot to watch the trailer and critique it, and the one who assembled the parts also failed to judge the video.

Talking about a 2D, you can ask for an awesome image and you publish then you realize it has 6 fingers lol... If only the artist had worked on it, removed a finger, and adjusted the hand.

AI is stealing art

Well, so do we. There is a quote that goes like this:

"Good artists copy, great artists steal"

To be a good artist, you will copy someone; to be great, we pick something that someone made and create something better, calling it ours. We are not re-inventing the wheel, we learn from other artist's works all the time, so why can't a programmer teach the machine to do the same? To me, a programmer is putting their back into making the AI learn from our work; that is impressive. Think about it, AI is not a crazy entity that appeared out of thin air; someone made it.

It's taking our job? I don't think so, at least not yet. If we think about it, every year new sculptors appear. Should we fear them?

If we want to talk about money, should we close Hero Forge? They commission artist and implement their work in the software to create thousands of miniatures. Are the artists being paid for every model sold? I don't know, but it's almost the same as AI to me.

Slop is fun

I had to include this. Sometimes, AI is such a nonsense slop that it can create some very funny videos that we as humans would never be able to do. Why? because we are bound by reality.

What I dislike

The thing that I dislike the most of AI, is the fact that it can mimic us and be used for harm. The AI is near a point that it can create videos of someone saying something they never did. Voices can be cloned to harass people. Can you imagine a fake kidnapping?

I fear this future.

Sorry for the long text, this topic is huge and has lots of views. I'm currently in neutral ground, tried to bring negative and positive views. I'm not asking you to agree or disagree with me. However, I ask you that try to see things for more than one perspective and be respectful regardless of your opinion. And remember that on the other end, there is a human (for now, bro, seriously, this is old school 80's trash movie)

Best regards,

Ivo

Files

Comments

Feltyr

I don't agree with your point. You cannot compare a human that goes through a process of learning, with a machine that they just drop petabytes of stolen content to copy. AI is a bubble, they make it free to show growing numbers, is just predictive text on steroids. There are too many enviromental and economical issues to just stay neutral and look the other way. Creating things is hard, and that what makes it valuable. People today have very short memory, but we had sh!tpost memes and post before the slop. At least there, someone went through the trouble to write a post or found the images for a meme. Using AI is just throwing money to a techbro so he can continue his ponzi scheme. They may try to shove AI everywhere, like they did with NFTs and so many other things before. But I don't have AI in my browser, and I disabled it on my phone. You can still ignore the slop result in google and go find an answer like you did two years ago. There is no positive side to generative AI. (Before someones points out: Yes, LLMs can be used in medicine and will probably save lives, but that's not generative AI. )

Imagine Minis

Yeah, sadly, they are pushing AI in everything everywhere, and kids that grow up in this world are a bit in trouble (even some adults). Using AI all the time, they forget to think for themselves. True People invested time doing memes all the time.

Jo Em

Huh, I wasn't prepared for something so long and well thought. Personally I use AI as a helping tool in my average day. I'm on parental leave until May 2026, so I don't have to *work* work with it, but the day might come. Usually, I use chatGPT for brainstorming or starting points, like "give me an idea for what I can cook with the ingredients XYZ" or a rough sketch idea for a paint scheme or stuff like that. I'm aware of the risks of wrong answers, which I avoid by not using AI for "serious" things. And I don't agree on the term "I put time and effort in something and that made it worth something". That's not true. I ditched countless projects, ideas, physical stuff and other things, which had already eaten time, money and efforts. The results were shitty and no one would've paid for it, even if this wasn't the intention of trying to achieve a good result. I think we are on a hype wave, that doesn't already reached it's peak, but it will. And observing and being aware of it is pretty damn important. I'm with you on the fear of harm that will be caused by AI, and I hope my kids will grow up with a serious amount of medial doubt, no matter if public information or personal communication. But I also think that AI won't leave, and hopefully humanity will force it to be a helpful tool, nothing more.

Adam Gibson

I just want to point out LLMs literally *ARE* genai. Gen AI is LITERALLY prompt -> output. ALL LLMs no matter how they're used *ARE* GENAI good or bad.You literally violated the definition of what these things are with handwaving. Like Ivo said they're two sides of the same coin. LLMs without chatgpt but LITERALLY just the bricks of math that they are are JUST something that converts different kinds of inputs to a vector it can understand run through a neural net JUST LIKE the object detectors you barely thought about before all this stuff became more powerful. We USED to laugh at this stuff 10 years ago when it was just encoders outputting cat pictures, now it's just a more powerful version of something that's existed for YEARs now.

Adam Gibson

Ivo: I just want to say I work in this space every day and am strongly conflicted about how the stuff is used. I STRONGLY believe in the good sides of these things while acknowledging how bad it is. Thank you for taking a more level headed stance on this stuff. I highly regret the slop and bad side effects coming from the abuse of the tools. As someone in the field I'm sorry about some of the BS that's come out of this. It's made the web simultaneously much worse but can REALLY help once people learn these things are bricks of math that can be abused just like any other tool. The regulations have DEFINITELY not caught up to the technology. I hope it does one day and we can start reducing the abuses of the tools being made.

Feltyr

No, LLMs are the underlying technology, Gen AI is what you do with it. So it's the other way around. If prompt -> output = AI. Then my very first exercise in programming printf("Hello, World!"); is AI. Your second paragraph is hard to understand, your predictive text must have changed a few words and it's lacking punctuation to separate terms. But my understanding is that you are saying that LLMs are just algorithms underneath that are more powerful than before. But in the past you only used them to recognize patterns, not GENERATE a regurgitated slop out of it. And the people that used them were just wanted to play with it at home or to experiment with it. They weren't megacorporations of techbros scouring the internet for copywrighted material in order to regurgitate slop that enables a brand new era of fake news. While consuming energy that sets us back in climate change goals. How you obtain and how the data is used matters. That is my whole last sentece is about. You can use it to detect cancer or find new treatments to save lives. Or you can use a tool that uses copywrighted material and enough energy for a small city to cheat exams, write CVs, create fake news, slop and whatever. No matter if you use it for something innocent. You are using the same tool and company that enables all the other terrible stuff. You using it is making their usage numbers better, and making the problem worse.