What AI-generated art really means for human creativity

picture Lee Unkrich, one of Pixar’s standout animators, as a seventh grader. He stares at the image of a locomotive on the screen of his school’s first computer. Wow, he thinks. However, some of the magic wears off when Lee learns the image didn’t just appear by asking for “a picture of a train.” Instead, it had to be painstakingly coded and rendered—by hard-working people.

Now imagine Lee 43 years later stumbling upon DALL-E, an artificial intelligence that creates original works of art based on human-supplied prompts that can literally be as simple as “a picture of a train”. As he types words to create one image at a time, the Impressive is back. Only this time it won’t go away. “It feels like a miracle,” he said says. “When the results came out, my breath caught and tears welled up in my eyes. It’s so magical.”

Our machines have crossed a threshold. All our lives we have been assured that computers are incapable of being truly creative. But suddenly, millions of people are using a new generation of AI to create stunning, never-before-seen images. Most of these users aren’t, like Lee Unkrich, professional artists, and that’s the point: they don’t have to be. Not everyone can write, direct and edit like an Oscar winner toy story 3 or coconutbut everyone can Launch an AI image generator and enter an idea. What appears on screen is amazing in its realism and level of detail. Hence the universal answer: Wow. On four services alone – Midjourney, Stable Diffusion, Artbreeder and DALL-E – people working with AIs create more than 20 million images every day. Brush in hand, artificial intelligence has become a wow engine.

Because these surprising AIs have learned their art from billions of images made by humans, their output revolves around what we expect from images. But because they’re an alien AI, fundamentally mysterious even to their creators, they restructure the new images in ways that probably no human would think of, filling in details that most of us don’t have the artistry for would have to imagine, let alone execute the skills. They can also be instructed to create more variations of something we like in seconds, in any style we want. Ultimately, this is their greatest advantage: they can create new things that are relatable and understandable, but at the same time completely unexpected.

In fact, these new AI-generated images are so unexpected that — in the quiet awe immediately following the Impressive– another thought comes to almost everyone who has met them: Man-made art must be over now. Who can match the speed, cheapness, size and, yes, wild creativity of these machines? Is art another human occupation that we must leave to robots? And the next obvious question: if computers can be creative, what else can they do that we’ve been told they can’t?

I’ve spent the last six months using AIs to create thousands of stunning images, often losing a night’s sleep in the endless search just one more Beauty hidden in the code. And after interviewing the developers, power users, and other early adopters of these generators, I can make a very clear prediction: Generative AI is going to change the way we design just about everything. Oh, and not a single human artist is going to lose their job because of this new technology.

it’s no Exaggeration to name images that were generated with the help of AI co-creations. The sobering secret of this new power is that the best applications are not the result of typing a single command prompt, but of very long conversations between man and machine. The progression for each image results from many, many iterations, back and forth, detours and hours, sometimes days of teamwork – all based on years of machine learning advances.

AI image generators are the result of combining two separate technologies. One was a historical set of deep learning neural networks that could produce coherent realistic images, and the other was a natural language model that could serve as an interface to the image engine. The two were combined into a voice-controlled image generator. Researchers searched the Internet for any image with adjacent text, e.g. B. captions, and used billions of these examples to connect visual shapes to words and words to shapes. With this new combination, human users could enter a series of words – the prompt – that describe the image they are looking for, and the prompt would generate an image based on those words.