Are you interested in getting started with digital art, NFTs, Artificial Intelligence (AI), or generative art?
Working with AI, Machine Learning (ML), and other digital tools are becoming more and more commonplace for artists as the tech has become easier and more ubiquitous.
In the guide, we’ll explain what AI art is, how it’s made, and some of the challenges and benefits of this type of tool as part of your creative practice. We’ll show you some examples and explain how this technology is changing the game for artists and hobbyists alike.
AI Art refers to any artwork created with the assistance of artificial intelligence. It can be a work created autonomously by AI systems or a work that is a collaboration between a human and an AI system.
Artists can use software to automate some part of their process with code – rather than working entirely by hand. Algorithms can be used to create everything from abstract works of digital art where lines are drawn according to a set of rules (you might be familiar with fractals) all the way up through using GANs to create a complete image.
Artificial Intelligence (AI) coupled with machine learning (ML) is all the rage these days. But what is it?
ML is often used for everything from facial recognition, ride navigation, and search engines but can also be applied to the world of art. This has led to some pretty amazing results. Artists are exploring new ways of creating work with this exciting new tool in their toolkit.
What is Machine Learning?
Machine Learning is a term that’s thrown around often these days. We hear it in the news, at work, and even on our favorite television show. But what does it mean? In traditional programming, a coder was required to write thousands of lines of code. Now a programmer can write some code that tells the program to ingest the internet and the program looks for patterns and writes itself. This is called Machine Learning (ML). The most important aspect of ML is the speed at which new programs can be written and deployed. ML can write programs much faster than traditional human-written code.
Wait… Algorithms Can Create Art?
Yep. Think about it this way: most artists have a process–or a set of processes–they use to create their work.
Any repeatable pattern is an algorithm.
Free Up Time for Higher-Level Work
Artists can free up time by automating some portion of the lower-level aspects of the creative practice and become more prolific in their work.
Introduce Unexpected/Random Elements to Encourage more Creativity
By introducing unexpected elements into the creative process, artists can stretch the boundaries of the imagination in ways that they would never have thought to previously. The ability to make something new, or ‘original,’ by factoring in random elements into their work. For example, someone could argue that there’s nothing really “new” about a landscape painting (at least at a general level). The artistry is in the act of making. But by introducing random elements into a creation that they might never have thought to otherwise can be introduced, they can make something truly novel. The use of aleatory in art has had wide usage with art groups, including Breton’s original French surrealist group, Oulipo, and the Beats.
Computers Are Just Another Tool
Like a paintbrush, spellcheck, or photoshop, an AI art generator app is just another tool to add to the artists’ toolkit.
Why use Algorithms to Create Art?
Just like folks don’t use typewriters anymore, there aren’t too many writers seeking to remove spellcheck from your computer. We believe that there is a place for using ML in any artist’s toolkit, no matter how classical their work may be.
Benefits of Using AI in Your Work
There are many benefits of using AI in the arts. Algorithmic art and generative artwork can create pieces at a speed that wouldn’t be possible manually. By iterating more quickly, artists can get those cliché, bad ideas out more quickly and spend long more time on those ideas that have more promise–ultimately creating better work.
What are the Different Kinds of Generative Art Methods You Can Use? Machine Learning, generative art (including fractals), glitch art, style transfer, GANs (Generative Adversarial Machine Learning), CANs (Creative Adversarial Machine Learning), and even literal robots (embodied AI) are some of the innovative tools artists and hobbyists alike are using to enhance their creative processes.
Genres of Digital Art
- Generative art: work created with rule-based pattern (this includes fractal shapes).
- Glitch Art: work created by using machine error
- Machine Art: robots as art
- Digital Painting: painting with a digital tool such as photoshop
- Machine Learning
- Style transfer
- Python Notebooks
- Physical Robots
There are infinite ways that AI-Generated art can be implemented into your creative practice. We’ll cover a few no-brainer use-cases below.
Background removal – Collagists, imaginative, digital artists, or surrealists can use background removal to easily collect image assets for their work by removing the backgrounds more quickly.
Object/person removal – Photographers and artists can easily use object/person removal to remove unwanted people and items from their images.
Random or algorithmic mixing and matching. Surrealist painters–or any creative folks for that matter–can quickly mix and match different objects, people, or landscapes to quickly iterate images ideas (tossing out the bad ones so that they can spend more time on the good ones.)
Upscaling and image sharpening – Photographers and hobbyists can use ML to sharpen blurry photos or to upscale their old photos to higher-quality, larger images.
Topaz Labs develops popular software for upscaling both photos and videos. Remini is another popular option.
Damaged Photo Healer – Fix damaged photos.
Animate old photos (or paintings) – Animate your Great Great Aunt Eliza’s photograph from 1920.
Colorize Old Photos – Old photo or black and white photo colorized. Old photo restoration with Photoshop.
Text prompt – AI can ideate images quickly. Painters can quickly iterate their sketches by inputting a text prompt, and the program can spit out image ideas/sketches.
Style transfer – Transfer styles using an art app to transform your picture into an oil painting with creative filters. Print the painting onto canvas, and then continue the oil painting. Make your photograph or art look like Van Gogh’s Starry Night. Cartoon yourself.
Robots – Using robotics to paint introduces random or unexpected elements in the work to force the creative process. If you don’t have a spare robot arm lying around, why not use a cheap robot vacuum?
Pixel Art – Turn a photo or sketch into pixel art for a more retro quality.
Turn a drawing, photo, or sketch into ASCII art
Animate children’s drawing – We’re sure there’s more to come, like Meta’s new tool here. Fun!
Cartoonify yourself – Turn your portraits into cartoon creations.
Grow the crop of a picture – Generate a larger image from an existing picture by growing the aspect ratio.
What’s an algorithm? An algorithm is just a sequence of step-by-step instructions. Think “paint-by-numbers.” An algorithm can range from very simple to extremely complicated. Any algorithm or computational task that a computer can accomplish, a human can achieve if given enough time.
What is algorithmic art? Also called code art or procedural art, Algorithmic art is typically visual art in which the design is generated by an algorithm. Fractal Art is a kind of algorithmic art.
Artificial Neural Network
Artificial neural networks, usually simply called neural networks, are computing systems inspired by the biological neural networks that constitute animal brains. An ANN is based on a collection of connected units or nodes called artificial neurons, which loosely model the neurons in a biological brain.
What is digital art? Digital art, computer art, or new media art refers to art made using software, computers, or other electronic devices. Digital art is an artistic work or practice that uses digital technology as part of the creative or presentation process.
Generative Adversarial Network (GAN)
A generative adversarial network (GAN) is an ML model in which two neural networks compete with each other to become more accurate in their predictions. GANs typically run unsupervised and use a cooperative zero-sum game framework to learn.
A GAN generative adversarial network is a class of machine learning frameworks designed by Ian Goodfellow and his colleagues in June 2014. Two neural networks are pitted one against the other (thus the “adversarial”) to generate new, synthetic instances of data that can pass for real data. They are used widely in image generation, video generation, and voice generation. with each other in a game. Given a training set, this technique learns to generate new data with the same statistics as the training set.
Creative Adversarial Network (CAN)
CANs are GANs that can think creatively.
Algorithmic art is a type of art that involves the use of algorithms to create or modify artwork. This can include anything from creating new paintings to remixing old ones.
I don’t think people even have a conception of what’s going to happen in their lifetime. Think about how technology is influencing our lives today. There will be even more variation in the future.
Generative Pre-trained Transformer 3 is an autoregressive language model that uses deep learning to produce human-like text. It is the third-generation language prediction model in the GPT-n series created by OpenAI, a San Francisco-based AI research laboratory. GPT-3 fuels Jasper. Jasper is quoted above and assisted in the creation of portions of this article.)
Convoluted Neural Network (CNN)
A convolutional neural network CNN is a class of artificial neural network most commonly applied to analyze visual imagery.
Glitch art is an aesthetic in digital art where the artist uses the malfunctioning of the tool intentionally. Methods can include opening a file in a text editor and typing random things before playing the media file again or an app like glitché. (“The Glitché iPhone App lets you corrupt and distort images to create your most psychedelic works of art”).
Embodied AI is a type of artificial intelligence that controls something in physical space –- a body, robot arm, Roomba, etc.
Fractals are infinitely complex patterns that are self-similar across different sizes. They are created by repeating a simple process over and over in an ongoing feedback loop. Driven by recursion, fractals are images of dynamic systems – the pictures of Chaos.
Turn any image into a Fractal Using Orbit Traps:
Interested in getting started playing around with AI artworks quickly? Try some of the below AI art apps! In our humble opinion, Dall·E 2 is currently the most versatile, with Midjourney a close second place… but both have their uses. We love Nightcafe for more abstract imagery and also video generation. For a slightly more technical audience, you might find Disco Diffusion Turbo with 3D output a great option for video as well.
Open AI’s DALL·E 2
DALL·E 2 is probably the most advanced AI art generator at this time. Unlike some of Open AI’s previous work, currently, the company has chosen not to open source its machine learning model citing safety concerns. DALL·E 2 can create photo-realistic generated outputs including faces. Because of this, Open AI doesn’t allow generating political art, nudity, or photographed faces to be uploaded due to the potential for fake news/imagery to be created and spread at an unprecedented rate.
Examples of Art Generated by Open AI’s DALL·E 2
Midjourney is a research lab and the name of the lab’s popular artificial intelligence program that creates images from textual descriptions. The tool is currently in open beta. The program was used by the British magazine The Economist to create the front cover for an issue in June 2022. [Source.]
Examples of Art Generated by Open AI’s DALL·E 2
Enter a text prompt, pick a style and watch the app create AI-generated artwork in seconds.
AI paintings from text prompt and image upload.
- Enter a prompt for the AI to work with (it can be anything! even emojis)
- Select a preferred style
Images above are upscaled with Gigapixel AI.
That’s it. Within minutes your AI-generated artwork will be ready. Free and paid option for iOS and Android.
Our new favorite: because of the ability to generate video! NightCafe Creator was founded by Angus Russell in November 2019 from the spare bedroom of his tiny semi-detached house in Sydney’s Inner-West. As of July 2021, over 300,000 AI-generated artworks have been created on the platform.
Angus started NightCafe after a house guest commented on how bare his walls looked. Angus agreed and jumped online the next day to look for some art to buy. After scrolling through hundreds of pages of artworks on every wall-art store he could find, he was left unsatisfied. Nothing seemed personal enough. You couldn’t have a conversation about it with a visitor.
Having known that AI Art had been a thing for years, he thought, “maybe I can generate some more personalized art using AI.” Surprisingly, his Google searches yielded few results. Forgetting about his own walls, Angus became obsessed with making this “obvious” idea available to the world.
Try NightCafe Creator Now
Get a 5% discount on NightCafe Creator with Coupon Code BECOME
Ganbreeder is a collaborative art tool for discovering images using BigGAN. Using the web app, you can mix or “breed” neural networks with each other into evocative, otherworldly imagery.
The results can then become physical art. Ganbreeder integrates with Ganvas Studio, which turns your GAN images into posters and paintings.
Runway is inventing the next generation of creativity tools. Runway is taking recent advancements in computer graphics and deep learning to push the boundaries of content and in turn, lower the barriers of content creation; unfastening a new wave of storytelling.
Runway is reimagining how we create; so we can create impossible things.
Best AI Photo and Video Enhancement
Not every photo or video requires exceptional image quality. For the ones that do, Topaz AI image enhancement software helps you achieve impossibly good results. Topaz Labs suite includes video and image upscaling as well as masking (for cutting out images in collages).
AI Photo Enhancer. Using movie-grade AI technology, Remini turns low-resolution, blurred, pixelated, old, and damaged photos into HD, with sharp and clear facial focus.
- Python (programming language) Notebooks
- Go (programming language)
Sophia Crespo’s Neural Zoo
Sofia Crespo’s work consists of different projects working with artificial intelligence, computed image recognition, and neural networks. Her project, Neural Zoo, explores how creativity combines known elements in a specific way in order to create something entirely new. In the process of generating new creatures that don’t exist yet, she offers a perspective on how similar human creativity works. The creator, in this case, would be the algorithm itself, but with a human artist as its muse.
Alexander Mordvintsev’s DeepDream
DeepDream is a computer vision program created by Google engineer Alexander Mordvintsev that uses a convolutional neural network to find and enhance patterns in images via algorithmic pareidolia, thus creating a dream-like psychedelic appearance in the deliberately over-processed images. Google’s program popularized the term (deep) “dreaming” to refer to the generation of images that produce desired activations in a trained deep network, and the term now refers to a collection of related approaches.
Mario Klingemann is a German artist and Google Arts and Culture resident known for his work involving neural networks, code, and algorithms. He is considered a pioneer in the use of computer learning in the arts.
Electric Sheep by Scott Draves
Scott Draves is the OG of digital artists. Draves created Electric Sheep in 1999. Electric Sheep is a form of artificial life: it is software that recreates the biological phenomena of evolution and reproduction through mathematics. The system is made up of man and machine, a cyborg mind with 450,000 participant computers and people all over the Internet.
Media artist, director, speaker, VJ. AI/ML, generative/interactive, new media art. Epstein’s style focuses on vivid figurative aesthetics at the junction of generative and figurative practices. The main interest in recent years is AI, new media, and creative coding.
Scott Eaton is an American artist, designer, and photographer. His work explores the representation of the human figure through various mediums – drawing, sculpture, photography, and generative AI.
In addition to his own work, Scott frequently collaborates with other artists and studios. Recent collaborations include work with legendary artist Jeff Koons, Turner prize-winner Mark Wallinger, Elton John, Nike, Pixar, Disney, and others. Scott Eaton
Gene Kogan is an artist and programmer with interests in generative art, collective intelligence, autonomous systems, and computer science.
Gene Kogan maintains ml4a.net, a free book about machine learning for artists, activists, and citizen scientists – and regularly publishes video lectures, writings, and tutorials to facilitate a greater public understanding of the subject.
Important considerations for using AI for artistic purposes
Like using any tool single tool in photoshop, the tool can quickly become recognizable and cliche. Consider that while your art can have familiar elements, simply using a single AI tool in your work can quickly feel cliche if the tool becomes popular. Consider using at least a few different tools to keep your work interesting and unique.
Is there a future for Artificial Intelligence Art in museums and galleries?
Yes, without a doubt. There already is.
What are the implications of artificial intelligence art on our society and culture in the future?
We believe that there will always be a place for all kinds of art, from classical representational to digital art and that for now, artists should consider AI or ML art as another kind of paintbrush or tool to work or play with or explore.
Artificial Intelligence is one of the most exciting technologies on the horizon. One way it can be leveraged now, however, is through art. We’ve provided some examples of AI artwork that have been created to give you an idea about what this type of work looks like. AI art has the potential to make the world more beautiful. It can also help us explore new ways of thinking about how we represent ourselves and our environment in visual form. While there are many benefits to using artificial intelligence in artistic endeavors (more time for creators, new creative opportunities), there are also challenges in using AI as a tool to create artistic works. What does AI art mean for human artists? How to address issues such as bias training? What is the original? How can working artists survive in a world where all goods are public goods? Can a piece be considered original if it was created without human input at all?
What do you think of AI art?