Crispydogs101 wrote:

You all guys are just wasting time ranting about something that's not even happening.

I mean you’re not wrong. The AI thing might be a proof of concept but at the same time they are talking about AI art. I’d rather speak out against it bow than wait until it’s already added to protest.

Crispydogs101 wrote:

You all guys are just wasting time ranting about something that's not even happening.

There is still a possibility of them adding it. They said they weren’t focusing on it right now, but they did say that they were looking at it, which brings the possibility for this to be implemented.

I feel that AI is just too new and unreliable for Scratch to use it. Maybe in a few decades.

BendyOl183 wrote:

Crispydogs101 wrote:

You all guys are just wasting time ranting about something that's not even happening.

There is still a possibility of them adding it. They said they weren’t focusing on it right now, but they did say that they were looking at it, which brings the possibility for this to be implemented.

I think they were just wanting to just see how it would look like. I think they know that it'll be very expensive to do so and won't even be able to add it.

Crispydogs101 wrote:

Okay I'm embarrassed.

Za-Chary wrote:

Strictly speaking, the Scratch Team never said that “they are considering adding AI image generation to Scratch.” All they said is that they are “exploring” it — that says nothing about adding it to Scratch.

That may be true, but I'd rather speak out against the idea while it's still in its early development than wait until the idea is implemented/well into development and then protest the idea.

Za-Chary wrote:

So the Scratch Team is already aware of the many problems with using AI image generation as part of Scratch — none of which are problems that you stated.

Yeah, I will admit that my original post was a little rushed. I have since updated it to go a little more in-depth. I also didn't realize that there was a blog post while making the OP.

Za-Chary wrote:

In fact, the Scratch Team seems to believe the opposite of what you list to be a problem — more possibilities are made using AI art, and the article explains how AI image generation can actually create more imagination. And besides, you do have to “imagine” something in order to tell the AI what art you want in the first place.

It requires almost no effort to tell an AI image generator that you want a red car. It does, however, take some skill to draw a simple red car.

Za-Chary wrote:

Scratch is a programming language, so it would be great if users could get started on the programming concepts as soon as possible while using sprites that they are happy with

Beginners might not notice it at first, but they don't have to worry too much about the art until their code is solid enough. Also, art skills can be improved over time if enough time is put into it.

Za-Chary wrote:

Things like AI image generation can also get users started with ideas (as the article states).

That's actually a decent point. No counterpoints for this one

Za-Chary wrote:

So I'd say your suggestion is already implemented, in a sense: the Scratch Team, at this time, is not considering adding AI image generation to Scratch. My guess as to why the Tweet was made in the first place is just to showcase the possibility of AI image generation within Scratch, and to showcase how cool it can be.

Yeah, I guess I was a little confused with the wording. I'm not going to close this topic though because I don't want AI art on Scratch to turn into reality (the chances might be low but they clearly aren't zero).

Yeah, sorry for not responding until now. IRL stuff happened

maybe they should close this topic?

BendyOl183 wrote:

There is still a possibility of them adding it. They said they weren’t focusing on it right now, but they did say that they were looking at it, which brings the possibility for this to be implemented.

The situation right now is basically like if you made an unshared (possibly inappropriate) Scratch project just to try something out, with no intent to share it on the Scratch website. The Scratch Team essentially doesn't care what you put in an unshared Scratch project, as long as you don't make any indications that you're going to be sharing this (possibly inappropriate) project on the Scratch website. Much like how we should not be outraged at the Scratch Team for simply exploring with AI image generation, given that they already said that they are not adding it to Scratch at this time.

ProjectFactory wrote:

That may be true, but I'd rather speak out against the idea while it's still in its early development than wait until the idea is implemented/well into development and then protest the idea.

I wouldn't really say it's “early in development,” or even “in development” at all — some Scratch Team members are just exploring with it (see my unshared project analogy above). I think it would be better to wait until the Scratch Team shows any indication that they are actually going to add this to Scratch — and, once that is the case, it may be better to focus on discussing how to improve such a feature rather than asking to remove it completely.

ProjectFactory wrote:

It requires almost no effort to tell an AI image generator that you want a red car. It does, however, take some skill to draw a simple red car.

Exactly. For those who don't have the skill to draw a simple red car, their imagination stops there and that potentially prevents a Scratch project from being made.

ProjectFactory wrote:

Beginners might not notice it at first, but they don't have to worry too much about the art until their code is solid enough. Also, art skills can be improved over time if enough time is put into it.

It would be nice if beginners could make projects that they like from the get-go, and part of this involves working with art that they are happy with. Considering that Scratch is primarily a coding platform rather than an art platform, it would make sense that a newcomer would prioritize learning how to code — and subsequently get discouraged if they code something that doesn't visually look good.

ProjectFactory wrote:

Yeah, I guess I was a little confused with the wording. I'm not going to close this topic though because I don't want AI art on Scratch to turn into reality (the chances might be low but they clearly aren't zero).

Why not discuss ways to make it better? If done correctly, I could see AI image generation as a great tool to help spark imagination among Scratchers. If AI image generation was perfect and had no problems as outlined in the article I linked, it would serve as a good tool. The problem you mention of “some users might become over reliant on AI to make art for them” might still be there, but again, one could argue just the same for things like Google Images, backpacking costumes, using the sprite library, etc…. I don't see anything wrong with Scratchers choosing not to draw their own sprites.

Uh actually I just found this.

google wrote:

even a small AI team can cost a business upwards of $320,000 per year in technology development costs alone.

And it says “small.” This is a huge financial issue if they were planning on adding it.

Crispydogs101 wrote:

You all guys are just wasting time ranting about something that's not even happening.

Yeah the post should probably close now

hydrofungus wrote:

Crispydogs101 wrote:

You all guys are just wasting time ranting about something that's not even happening.

Yeah the post should probably close now

I agree.

This is turning into the “GRRR I hate AI” topic.

I've already dealt with those people on other sites(One AI fan stated how they kept recieving nasty comments from human artists, and a guy that made a Scooby Doo animation with AI voices(because shoestring budget) was put on blast by a professional voice actor), I don't want the same to happen here.

Adding AI(something REVILED by artists, and a frequent cause of drama in the art community) to Scratch(A site designs to stifle hate and drama) is a recipe for disaster.

I think this topic has come to a proper conclusion that AI art won't be added soon. Therefore, I've closed it myself.