Slay the Spire 2 lead says the garbage placeholder art is "important" and gen AI would ruin it: "People would judge the game as if it's a 1.0, fully developed video game"
AI-generated art looks "very strange," says Casey Yano
Slay the Spire 2 devs are purposely choosing to use delightfully "shit" human-made placeholder art in their roguelike instead of AI-generated assets, the game's lead says – not just because the latter looks "lazy," but because "it's important" that the unfinished art looks distinctly, well, unfinished.
Speaking to GameSpot in a new interview, Mega Crit co-founder Casey Yano discusses Slay the Spire 2's approach to placeholder art, which you can find a bunch of in the game's Timeline. And, seriously, you can't miss it – I mean it with all the love in my heart when I say these artworks look like they were created in MS Paint, but according to Yano, this is very intentional.
This is largely because Slay the Spire 2 is still in early access, and Yano "wanted to set the precedent that some of the stuff is incomplete, and I wanted to make it obvious." Even though he reckons the roguelike's artists might have wanted more of the game's art to be complete by the time the initial launch arrived in March, Yano "knew that we were going to be drawing a bunch of garbage during early access," and believes that the incomplete look is "important" since "otherwise people would judge the game as if it's a 1.0, fully developed video game."
He continues: "If we use art that looks nearly complete, then people would think that that's going to be the final art. It has to look like shit. It's important that it looks like shit."
Of course, the placeholder art still serves an important purpose. Yano wants it to serve "kind of like a framework, especially – especially – the art that's used in the timeline." After all, those are "the narrative beats," and the lead wants to ensure that their planned compositions, as well as details and colors, are properly "communicated through those drawings." The devs don't let just any placeholder art make it into the public version of the game, either, as Yano discusses one artist on the team known as Jose who "tends to draw a lot of inspiration from pop culture."
He continues, jokingly: "It's really bad. We're [going to] get sued by Nintendo soon. He's no longer allowed to draw placeholder art now that we're in early access and it's more public-facing. We have some suspicious things in there. Some Among Us, some SpongeBob."
Beyond all that, though, Yano takes a stance against AI-generated art. "Anybody who has gone through the [artistic] process… I wouldn't say they would feel cheated, but they would feel a kind of sadness, right? You don't see the path that somebody took to get better and develop their own style. You just see somebody who's like, 'I just want something in this style,'" he explains. "And there are a lot of artists out there that make works that are meant to imitate certain styles. There's a lot of jobs where we imitate different things. But in this case, the game is in our style, so I don't know why we'd need AI to mimic our own style."
Weekly digests, tales from the communities you love, and more
Not only that, but with human-created art, "there's a decision process for nearly every line – every line you erase, every line you adjust, and everything that you do." Meanwhile, AI-generated art has "a very distinct look" that Yano finds "very strange." He says: "It's very dense, very colorful, and very distinct. It doesn't fit with the rest of the art style and it's just clearly lazy."
He's not wrong about AI-generated art standing out like a sore thumb – we've now seen backlash many times over different games after fans have spotted something distinctly non-human lurking under the surface. For example, Crimson Desert launched with a number of AI-generated paintings, which Pearl Abyss swiftly apologized for – stating that the "intention has always been for any such assets to be replaced" – before going on to do so in a following update.
Similarly, 2025's J'RPG sweetheart Clair Obscur: Expedition 33 was accused of including an AI-generated asset of some newspapers, which Sandfall quietly changed in an update. However, later in the year, The Indie Game Awards revoked the game's Game of the Year and Debut Game award wins, saying: "When it was submitted for consideration, a representative of Sandfall Interactive agreed that no gen AI was used in the development of Clair Obscur: Expedition 33. In light of Sandfall Interactive confirming the use of gen AI art in production on the day of the Indie Game Awards 2025 premiere, this does disqualify Clair Obscur: Expedition 33 from its nomination."
In the case of Slay the Spire 2 developer Mega Crit, Yano isn't dead set against AI as a whole, calling it "like a glorified spell checker" for some of the studio's programmers who use it for code reviews. However, generative AI is a different story, and something that the team stays clear of – as Yano puts it, art is "an expression," not something that has "an end goal that's definable."
Why so many game developers don't want to use generative AI.

I'm GamesRadar+'s Deputy News Editor, working alongside the rest of the news team to deliver cool gaming stories that we love. After spending more hours than I can count filling The University of Sheffield's student newspaper with Pokemon and indie game content, and picking up a degree in Journalism Studies, I started my career at GAMINGbible where I worked as a journalist for over a year and a half. I then became TechRadar Gaming's news writer, where I sourced stories and wrote about all sorts of intriguing topics. In my spare time, you're sure to find me on my Nintendo Switch or PS5 playing through story-driven RPGs like Xenoblade Chronicles and Persona 5 Royal, nuzlocking old Pokemon games, or going for a Victory Royale in Fortnite.
You must confirm your public display name before commenting
Please logout and then login again, you will then be prompted to enter your display name.
