The programs of artificial intelligence (AI) that allow you to create images from written descriptions are becoming valuable creative tools.
During the last months, users have been able to be surprised (and delighted) with the creations of Craiyon, DALL-E 2, Midjourney and Stable Diffusion, to name a few, but now we are just beginning to see more complex use.
And there are people who have begun to combine the power of these AI imaging programs with other tools to do a great job.
This is the case of William Buchwalter, a former engineering engineer machine learning from Microsoft Research that used the compositing capabilities of Diffusion Stabilization with a plugin of Photoshop called Alpaca to create an amazing “digital paint” using his computer.
AI and design tools combined to unleash creativity
Buchwalter is not a digital designer or illustrator. In fact, he admits that he doesn’t know how to use a graphics tablet. However, he seems to have a special skill for image generator you chose to create exactly what you wanted.
Instead of drawing “a house on a hill”, the young man asked Stable Diffusion. He does the same with “a small farm” and other things you imagine.
When the imager is running, Buchwalter is responsible for place the images on different parts of the Photoshop canvas and, using your editing skills, you are responsible for combining them.
Result? A completely new creation that human abilities or the Stable Diffusion AI itself could not independently create.
On the other hand, the artist Martin Nebelong He is also interested in exploring and exploiting the capabilities of next-generation AI-based tools.
In his case, he applies drawing techniques, but also uses Stable Diffusion, albeit with a collaborative art background called artbreeder. Some of his most recent work can be seen on his profile. Twitterand can be seen by entering this link.
“How to generate an owl? AI imaging tools are rapidly evolving these days, and this is one of the latest. The new beta version of Artbreeder is powered by Stable Diffusion, and it is a different version of what tools like this could become”, can be read in the tweet translated into Spanish.
With these kinds of possibilities coming up strongly, it’s good to keep in mind that these are just a few examples. Yes, you are new technologies open debates with both positive and negative opinions.
There are those who support it and say that the new tools will help democratize visual storytelling, while others regret having learned the art, perhaps fearing that in the future machines will do their work for them.
Images may or may not be copyrighted
But that is not the only dilemma with this question. An important point that is also explored refers to the copyright of the images. The AIs used to create the images were trained on large data sets from the internet, which they can include the work of professional photographers and digital illustrators.
Are you facing copyright issues? What about the combined images? The World Intellectual Property Organization (WIPO) reminds that only humans can be authors of creative works.
Then, since the AI is not a person, there is no wrongdoing. However, these organizations plan to adapt their standards to these new times as imagers take on an increasingly prominent role.
Over time, it will be possible to know what role these imaging systems will play and how they will become the standard for future artistic creations. The truth is that artificial intelligence is growing by leaps and bounds.