--- title: "We’re Cry-Laughing at These “Spectacular Failures” of AI Generated Art" description: "AI image generators are not perfect as a channel called \"failed-diffusions\" on Stable Diffusion's public Discord server goes to show." date: "2022-10-24" modified: "2022-10-24" authors: - name: "Victor Tangermann" job_title: "Senior Editor" link: "https://futurism.com/authors/victor" url: "https://futurism.com/the-byte/ai-generated-art-failures" categories: - "Artificial Intelligence" tags: - "artificial intelligence" - "image generator" - "stable diffusion" - "the digest" --- # We’re Cry-Laughing at These “Spectacular Failures” of AI Generated Art ![AI image generators are not perfect as a channel called "failed-diffusions" on Stable Diffusion's public Discord server goes to show.]() *\Image: Stable Diffusion\* ## AI Fails AI image generators like DALL-E2, Stable Diffusion, and Midjourney have [come a long way](), blazing a new path towards the ultimate freedom of artistic expression. In a shockingly short period of time, they've allowing practically anybody with an internet connection and a keyboard to generate some incredible, photorealistic art from simple text inputs. But they're not perfect. As a channel called "failed-diffusions" on Stable Diffusion's public Discord server goes to show, these image generators don't always hit the mark, with often hysterical results. Images [shared by American technologist Andy Baio](), which have since gone viral, demonstrate that Stable Diffusion isn't yet fool proof — or, at least, that it's much less stable than you'd think from the name. https://twitter.com/waxpancake/status/1583879929472897024 ## Nightmarish Animals From misplaced eyes and arms to nightmarish monstrosities from the AI-inspired animal kingdom, we can't help but get a chuckle out of these. https://twitter.com/waxpancake/status/1583879934388625409 When it comes to generating images of hands and feet, Stable Diffusion is clearly having some issues. https://twitter.com/waxpancake/status/1583925377118130178?s=20&t=7UtzLqiN6Cnsft4Feeq5Dw Other images show the algorithm simply failing to get a grasp on where one object begins, and another ends. ![]() ![]() ![]() ## Assuming Control All of this isn't to say that image generators like Stable Diffusion aren't powerful. As we've seen, these algorithm-based tools have become incredibly good at generating photorealistic art that can not only fool the eye, but [even mislead as well](). Given the current state of things, as evidenced by the examples above, we've still got some time before AI generators can truly take over the world — and until that happens, we might as well have a laugh at some early attempts. **More on AI image generators:** *[New AI Tool Turns Your Ugly Photos Into a Beautiful Selfie]()* ## Author I've been at Futurism since 2017, where my role has evolved to encompass design, writing, and increasingly editing. I've always been fascinated by space exploration and advanced transportation, which I've leaned into by interviewing luminaries in those fields while closely following the dimensions of policy and regulation that allow next-generation projects to succeed -- or, sometimes, to fail. I'm also keenly interested in the effects of generative AI on society, policies, and democratic institutions, as well as clean energy, physics and biology, and the vagaries of tech leadership. My work for Futurism has been cited by publications including Ars Technica, Gizmodo, PC Magazine, Jalopnik, Fox News, and the New York Post. I spent my childhood living in locations including Manila, the Philippines, and Geneva, Switzerland, attended McGill University, and now live in Toronto, Canada. Before Futurism I worked at AskMen and a small photography studio. In my free time, I'm an avid gardener, foodie, and craft beer lover, as well as a maker of artisanal hot pepper sauces. I have a magnificent dog named Freida. ### Author social links [Bluesky]()