A picture is worth a thousand words, but due to an artificial intelligence application called DALL-E 2, you can have a professional-looking image with many less.
DALL-E 2 is a new way to use neural networks that lets you make an image from a short word or sentence. The initiative, which was launched by the artificial intelligence research center OpenAI in April 2022, has not yet been made public. But a small but growing number of people, including myself, have been given permission to try it out.
As a scholar who is interested in how technology and art interact, I was excited to see how well the app worked. After hours of testing, it’s evident that DALL-E, although not without flaws, is light years ahead of current picture generating technologies. It raises very important questions about how new technologies will change how art is made and how people experience it. It also raises issues about what it means to be creative when DALL-E 2 seems to automate so much of the creative process itself.
Instead of using photos from the web, the computer generates six entirely unique images, each of which reflects some variation of the input word. (Until recently, the application created 10 photos for every prompt.) For example, when my friends and I gave DALL-E 2 the word prompt “cats in devo hats,” it generated ten photos in various styles.
Almost all of these might pass for professional shots or drawings. While the algorithm could not fully comprehend “Devo hats,” the unusual helmets used by the New Wave band Devo, the headgear in the photos it created came close.
A tiny group of artists has been employing neural network algorithms to create art during the last several years. Many of these paintings have unique features that nearly appear like actual photographs, but with strange distortions of space—a kind of cyberpunk Cubism. The most contemporary text-to-image technologies often generate hazy, surreal images that may be enjoyable but seldom appears genuine.
DALL-E 2 provides a considerable improvement in visual quality and realism. It may also imitate certain styles with astonishing precision. It will create six life-like photos if you desire images that seem like genuine photographs. If you desire ancient cave paintings of Shrek, it will create six images of Shrek that seem like they were made by a prehistoric artist.
It’s mind-boggling that an algorithm can achieve this. Each picture collection is created in about a minute. Not all of the photos will be attractive to the eye, nor will they necessarily match what you had in mind. Even with the necessity to filter through multiple outputs or test other text prompts, there is no other method to produce so many amazing outcomes so rapidly – not even employing an artist. And sometimes the unexpected outcomes are the finest.
In theory, anybody with sufficient money and ability may build a system like this. HuggingFace, a company, is openly creating its own version that anybody can test right now on the web, however it’s not yet as good as DALL-E or Google’s system.
It’s easy to see how these technologies could change the way people make pictures and talk to each other, whether through memes, greeting cards, advertising, or even art.