WotC: 'We made a mistake when we said an image not AI'

It seems like AI art is going to be a recurring news theme this year. While this is Magic: the Gathering news rather than D&D or TTRPG news, WotC and AI art has been a hot topic a few times recently.

When MtG community members observed that a promotional image looked like it was made with AI, WotC denied that was the case, saying in a now-deleted tweet "We understand confusion by fans given the style being different than card art, but we stand by our previous statement. This art was created by humans and not AI."

However, they have just reversed their position and admitted that the art was, indeed, made with the help of AI tools.

Well, we made a mistake earlier when we said that a marketing image we posted was not created using AI. Read on for more.

As you, our diligent community pointed out, it looks like some AI components that are now popping up in industry standard tools like Photoshop crept into our marketing creative, even if a human did the work to create the overall image.

While the art came from a vendor, it’s on us to make sure that we are living up to our promise to support the amazing human ingenuity that makes Magic great.

We already made clear that we require artists, writers, and creatives contributing to the Magic TCG to refrain from using AI generative tools to create final Magic products.

Now we’re evaluating how we work with vendors on creative beyond our products – like these marketing images – to make sure that we are living up to those values.

This comes shortly after a different controversy when a YouTube accused them (falsely in this case) of using AI on a D&D promotional image, after which WotC reiterated that "We require artists, writers, and creatives contributing to the D&D TTRPG to refrain from using AI generative tools to create final D&D products."

The AI art tool Midjourney is being sued in California right now by three Magic: The Gathering artists who determined that theirs and nearly 6,000 other artists' work had been scraped without permission. That case is ongoing.

Various tools and online platforms are now incorporating AI into their processes. AI options are appearing on stock art sites like Shutterstock, and creative design platforms like Canva are now offering AI. Moreover, tools within applications like Photoshop are starting to draw on AI, with the software intelligently filling spaces where objects are removed and so on. As time goes on, AI is going to creep into more and more of the creative processes used by artists, writers, and video-makers.

 

[QuentinGeorge]

Sure, like Adobe did. But then, what's the problem with the example by WotC? They mention specifically that no-AI policy is made more difficult to enforce with its integration into Photoshop, but if the opposition to AI is only linked to the legality of the acquisition of data, there should be no problem in keeping the "contested" pieces? I think there are two distinct things: first, there is the transient problem of the uncertain legality of datamining over the Internet (transient because it will be settled by lawmakers at some point and because there will be enough public domain images to build excellent models on, or GAN will be made to train AI models on AI-generated images) and the more fundamental problem of being against AI on principles. Those two shouldn't be confused.

I actually think the photoshop situtation here is silly and an overreaction. An AI-tool in a graphic design program that is not trained on copyrighted data is 100% ok in my mind. No different to someone writing using a grammar/spelling checker or something like AutoCrit/Grammarly.

 

[Jfdlsjfd]

It's a flawed analogy.

Tide = AI.
God = Corporations and Billionaire Class.
King = Those saying it's impossible.
Courtiers = Those saying that it is possible.

The irony being, it could be stopped, because it's not some higher power in control. It's just us, people.

Human nature is certainly a higher power. People always ended up choosing to improve their general quality of life against the interest of the few. Very few people support tailors by buying bespoke clothings and patching them instead of having large wardrobes and discarding nearly-new ready-to-wear items, very few people support postmen by sending letters instead of emails for non-urgent communications, very few people choose to buy local when there is an equal offering from a lower-wage country... There will be some who will prefer to commission an artist over drawing an AI image of their PC to put on the character sheet, but I guess in time there will be as few as the above categories. Especially among the large part of the population who so far elected to not have images at all because they didn't want to pay the price of commissionned art and for which having the zero cost opportunity, or nearly zero, since while generative AI software can be had for free, there is the need of computer, it would be a net improvement of their QoL.

 

[dragoner]

Anyway, apparently OpenAI said it would be 'impossible' to train if not able to use copyrighted materials.

I'm sure the billionaires wept.

One could see storm clouds on the horizon when AI art couldn't be copyrighted, they could tell what was coming. Part of this is that the release and patch business model doesn't work here. Even the fair use argument is pretty thin as it is not used by businesses in general, afaik.

 

[StarkMaximum]

AI will continue to be a news bullet so long as companies continue to not just use AI, but continue to lie to their consumers about it being AI in the hopes that we can be tricked.

 

[teitan]

The fix is easy, return to physical media.

 

[MGibster]

But I doubt we'll reach a point where, beyond some basic imagery, we'll stop being able to tell human art and AI "art" apart, because the AI's output will always be lacking in express. It will always lack soul. And thus, what AI produces is not art at all; subpar or otherwise.

So what you're trying to say is that one day it will be impossible to tell the difference between Thomas Kinkade's work and AI?

 

[trappedslider]

The fix is easy, return to physical media.

 

[Art Waring]

It is both easy and worthwhile to target AI "art" for the things it does incredibly poorly; faces and hands tend to be the most common "gotchas", as well any kind of written text. But the problems go deeper: you mention the proportions and composition, of course, which goes into AI's inability to understand depth of field and... well... yeah, composition. But the problems go deeper still. And here I'm not even touching on the thornier ethical issues concerning "training data" or artists losing work, which are also both incredibly important.

Yeah I agree, I was basically just responding quickly to the post to address the most surface-level problems I had with the images.

I have said as much regarding the deeper issues in previous posts in the past months, such as the problematic nature of the LAION 5B training dataset (starting as a non-profit dataset that is now being used for commercial purposes), and the lack of artist consent, or opt-in instead of opt-out options for artists, and other issues.

Much of it has largely fallen on deaf ears. So, I kind of stopped talking about it, especially when a lot of folks are still cool with all of it. If that's what they want to do, who I am to tell them otherwise.

Writing graff in the dead of night, headphones blasting out industrial rhythm laced with drum & bass, feeling a moment of freedom inside lines painted in red and black.

All those moments will be lost in time, like... tears in rain. Time to die.

 

[Vincent55]

But it's not a good analogy. Most automated work is done because it saves labour, and the machine can do the task without us. But AI-generated art literally cannot function without a training dataset. It is dependent on the creative output of humans which it then mindlessly copies. It's literally like a robot JJ Abrams.

Well, it is a good one because no computer can function without creative input from those who write the base code, it took creative people and very smart ones to write the code. This is no different than that, also you can't copy write an idea or feel or even style or look, well not indefinitely that is. Tolkien, Edgar Rice Burroughs, Robert Ervin Howard, and many more works and others are the base that others built off of as well as art from many others like frank Frazetta, Larry Elmore, Todd Wills Lock, Brom and others. Their style is copied by other artists and people around the world, are they cheating and stealing from them or are they being inspired by them? As i said doesn't matter to me and in the end, we are not going to stop them from using it, good or bad it comes down to money and the cheaper ways are almost always going to overshadow the more expensive slower ones. I do agree i prefer an artists version over some AI but i use the AI to inspire me to create my ideas which are more specific and less a random generalization like A.I. is, so in that aspect, it may never be 100% spot on well until it becomes self-aware that is, then we will be back to having to compensate them for their work and giving ai credit for its work. But I am not arguing for its use just saying we are not going to stop it and it will just come to a point when we will just not know it is being used, and i would bet it is in many areas we just don't know it.

 

[teitan]

Then people will keep saying everything is AI because everyone is an expert on the internet.