November 03, 2022
DALL-E: a tool that draws anything
Discover DALL-E, the famous new Artificial Intelligence technology which is able to transform a description of an image into a real image comes from two words: “WALL-E,” the 2008 animated movie about an autonomous robot, and "DALI" - a reference to Salvador Dalí, the surrealist painter.
DALL-E was created by OpenAI, a seven-year-old San Francisco company whose mission is to create safe and useful artificial general intelligence. OpenAI is already well known in its field for creating GPT-3, a powerful tool for generating sophisticated text passages from simple prompts. And now it's time for the images. DALL-E doesn't only create images but also, does the editing part. Let's imagine you want to generate an image of a mug on a table in the middle of the forest. However, a minute later, you realise that you need a candelabra instead of a mug. Simply erasing one object and substituting it with another is absolutely possible.
A team of seven researchers spent two years developing the technology. Now it became an excellent tool for all type of artists providing new shortcuts and creative ideas as they generate and edit digital images.
DALL-E is what artificial intelligence researchers call a neural network, which is a mathematical system loosely modelled on the network of neurones in the brain. That is the same technology that recognises the commands spoken into smartphones and identifies the presence of pedestrians as self-driving cars navigate city streets.
Though DALL-E often fails to understand what someone has described and sometimes mangles the image it produces, OpenAI continues to improve the technology. Researchers can often refine the skills of a neural network by feeding it even larger amounts of data. Therefore it means that the more images of mugs, candelabras and forests DALLE-E is exposed to, the more perfect the final result of the visual representation will be.
Like in every system, tool or application, there are rules for requesters. Images of public figures are off limits, as is anything remotely offensive, including nudity and violence. Political campaigning is forbidden too.
For creation of these images, 1605 Collective used the following keywords in the search: charcoal, black-and-white, Cubism, Minimalism,
Ukiyo-e, Abstract Expressionism.