Image Lee Unkrich, one of Pixar’s most famous animators, as a seventh grader. He was staring at the locomotive on his school’s first computer screen. Oh, he thought. However, some of the magic disappeared when Lee learned that the image did not appear simply by asking for a “picture of a train”. Instead, it has to be meticulously coded and rendered—by hard-working people.
Now imagine Lee 43 years later, stumbling upon DALL-E, an artificial intelligence that generates original works of art based on human-supplied cues, which can be as simple as “pictures” about a train”. As he typed words to create image after image, oh has returned. Only this time, it didn’t go away. He feels like a miracle, he speak. “When I got the results, I was choking and had tears in my eyes. So magical.”
Our machine has passed a threshold. Throughout our lives, we have been assured that computers are not capable of truly creation. However, suddenly, millions of people are now using a new type of AI to create stunning photos like never before. Most of these users are not professional artists, like Lee Unkrich, and that’s the point: They don’t have to be. Not everyone can write, direct, and edit an Oscar winner like Toy Story 3 or coconutbut everyone maybe launch the AI image maker and enter an idea. What appears on the screen is amazing in terms of realism and depth of detail. Hence, the universal response: Oh. Across four services alone—Midjourney, Stable Diffusion, Artbreeder, and DALL-E—humans working with AI currently co-create more than 20 million images per day. With a paintbrush in hand, artificial intelligence has become an engine of wow.
Because these surprise AIs have learned their art from billions of human-made photos, their output revolves around what we expect the photos to look like. . But because they’re an alien AI that’s essentially mysterious even to their creators, they reconstruct new pictures in ways no human could ever think of, filling in the details. details that most of us wouldn’t have the artistic ability to imagine, let alone the skills to execute. They can also be instructed to create more variations of something we like, in any style we want—in seconds. In the end, this is their strongest advantage: They can create new things that are relevant and understandable but at the same time completely unexpected.
In fact, these new AI-generated images were so surprising that—in the silent amazement that followed oh—another thought that comes to most people who have encountered them: Man-made art must now come to an end. Who can compete with the speed, cheapness, scale and, yes, frenzied creativity of these machines? Is art yet another human goal that we must yield to robots? And the obvious next question: If computers can be creative, what else can they do that we’re supposed to be unable to?
I’ve spent the past six months using AI to create thousands of outstanding photos, often losing sleep in my restless search. just one more beauty hidden in the code. And after interviewing the creators, power users, and early adopters of these generators, I can make a very clear prediction: Innovative AI will change the way we design plan everything. Oh, and not a single artist has lost a job because of this new technology.
it’s not exaggeration to call the image created with the help of AI co-creation. The sobering secret of this new power is that its best uses are not the result of typing a single prompt but of very long conversations between man and machine. Progress for each image comes from lots of iterations, back and forth, detours, and hours, sometimes days, of teamwork—all thanks to the company’s years of progress. machine learning.
The AI image generator was born from the combination of two distinct technologies. One is the history of deep learning neural networks that can produce consistent realistic images, and the other is a natural language model that can serve as an interface to an imaging engine. The two are combined into a language-based image generator. The researchers searched the internet for all images with adjacent text, such as captions, and used billions of these examples to connect visual forms to words and words to expressions. sample. With this new combination, users can enter a string of words—reminder—describing the image they’re looking for, and the prompt will generate an image based on those words.
