Artificial Intelligence Art refers to any artwork created with the assistance of AI. It includes works created autonomously by AI systems and works that are a collaboration between a human and AI system.
To create AI art, artists write algorithms not to follow a set of rules, but to “learn” a specific aesthetic by analyzing thousands of images. The algorithm then tries to generate new images in adherence to the aesthetics it has learned.
In October 2018, this “art piece”: Portrait of Edmond de Belamy, an algorithm-generated print, was sold for $432,500, thus beginning the AI-Art goldRush.
Price last sold USD 432,500
Listed Price Estimate USD 7,000 – USD 10,000
Closed: 24 Oct 2018
The world’s largest community of artists exploring Artificial Intelligence. https://aiartists.org/
It includes the list of top 25 AI Art Artist in the field and their contribution toward Art at large.
Examples of AI Art
Prominent artists using AI in their creative practice include:
- Refik Anadol – Architecture and AI
- Sougwen Chung – Drawing and AI
- Mario Klingemann – Animation and AI
- Mauro Martino – Sculpture and AI
- Lauren McCarthy – Performance and AI
- Wayne McGregor – Choreography and AI
- Trevor Paglen – Creativity and AI
- Anna Ridler – Photography, video and AI
- Pindar Van Arman – Creativity and AI
>>> Artwork generated using VQGAN and OpenAI’s CLIP
In a recent exhibition of prints shown at the HG Contemporary gallery in Chelsea, the epicenter of New York’s contemporary art world, 20 prints were displayed as part of the “Faceless Portraits Transcending Time.”
Tools and processes
There are many mechanisms for creating AI art, including procedural ‘rule-based’ generation of images using mathematical patterns, algorithms which simulate brush strokes and other painted effects, and artificial intelligence or deep learning algorithms such as generative adversarial networks and transformers.
Since their design in 2014, Generative Adversarial Networks (GANs) are often used by AI artists. This is an AI system consisting of two elements. The first is the “Generator”, the second is the “Discriminator”. The discriminator is fed with images of existing art works, whilst the “Generator” attempts to create new images. The role of the “Discriminator” is to see if it can spot which works are computer generated, whilst the generator tries to create works which fool the discriminator. If the discriminator fails to spot a piece, that piece is considered successful.
DeepDream, released by Google in 2015, is one of the more well-known AI art tools. DeepDream 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.
OpenAI released a series of images in January 2021 created using one of its algorithms “DALL-E”. The program can use AI to generate a variety of drawings and pictures based on various text prompts.
Here is an interesting video from Wired.
Computers as artists: Artbreeder
AI technologies opened a new door for photography and painting: mixed images with AI. These AI platforms are a collaborative art tool to discover and create new images based on the concept of “children” image that are made from other, user-generated mixes.
Artbreeder, for example, is a powerful tool for creators to generate inspiring images based on collaborative, children images. It can create portraits, album covers, landscapes, and more. Artbreeder is based on the Generative Adversarial Networks and the BigGAN models. This form of deep learning has been used in other programs like Artbreeder, like Google’s well known Deep Dream Generator. Artists flout this technology as new, unique tools for creators.
Artbreeder is another example of how open source, collaborative programs are vital for innovation. These projects allow artists and programmers alike to create art of the future.