Impossibly realistic and creative art created by artificial intelligence has been appearing more and more frequently over the past few months, and now anyone can run a full graphical version of the text-to- image artificial intelligence Stable Diffusion on any Apple Silicon- You just have to put the app in the Applications folder and write your prompt. You can download it here.
Until now, there were only two options if you wanted to take part in text-to-image artificial intelligence. If you were lucky enough to get an invite, you can pay to use a service like MidJourney or DALL-E. While there are step-by-step guides to install all the required components, they miss a few things that make it frustrating.
Thanks to Divam Gupta, all that has stopped. Stable Diffusion is easy to install on an Apple Silicon machine. A simple drag-and-drop is all that is needed. It doesn't need a connection to the internet beyond the initial download of the models needed to make it work. Information is not uploaded to the cloud. It is possible to run it on any Mac with an M1 or M2 processor. Even though machine learning requires a lot of memory, I have been able to run it on a MacBook Air M1 with 8 gigabytes of RAM. One tip is to close all your apps to give Diffusion Bee as much space as possible.
It is a remarkable development that will cause an explosion of new images and creativity with consequences that we can't yet see.
While many illustrators have lambasted these image-to-text tools, most of the time out of fear of losing their jobs, there is a growing feeling that this kind of software will be just another tool in the arsenal of creatives.
Video artist and director Paul Trillo told me a couple of weeks ago that he believed this tool wouldn't take any jobs away from visual-effects artists. He expects that it will create efficiency to the work they are already doing. It will allow for lower-budget projects to have realistic visual effects.
Josephine Miller told me that the technology was allowing her to do more things. She says that sometimes she feeds her designs to DALL-E, which produces variations of it, and then she discovers something unexpected that she didn't think about. Miller says that she has used the augmented reality filters to present variations of her work to her clients. Miller says, "I tell them this is my design, but these others were created by the artificial intelligence for you to see."
Sainsily is an artist and a design manager at Unity. He told me that they open a path for people with no visual execution skills to create something visual. He says it can help people who have no power. The case in which disabled kids were suddenly able to use their words to create images using DALL-E is one that Miller agrees with. She says it was wonderful.
Sainsily thinks that this technology will lead to a renaissance like we have seen before with other technical revolutions, like digital video editing, desktop publishing, or photography. For hundreds of years, we have been tweaking. The process is made quicker by this technology. It will bring a lot of change to the industry but it will also offer great opportunities.
This creative moment is inevitable even if you don't agree with Stable Diffusion being a one-click app. It will become increasingly difficult to control the use of sampling because of the laws and lawsuits that will be put in place. Right now, visual artists process bits and pieces of other peoples work to create everything, from storyboards to full artwork, without crediting the original samples. Many visual industries already reuse other people's work, so it's hard to blame artificial intelligence for doing that. People are constantly creating new things. Sainsily says that the artificial intelligence makes it quicker.
Current industry practices and the nature of the artificial intelligence tools will make any regulation fall flat. With time, it will become truly synthetic and erase any trace of the original work. We can use this early version of the technology to our advantage.
All Rights Reserved © 2024
