Wordsmiths and photographers are somewhat freaked out about AI, and for good reason, as the internet is destined to be saturated with AI-generated content. We are already seeing creations created by AI Amazon Kindle Books And photo contest winning images. And given that, for various reasons, many humans who use AI prefer not to admit to having done so, how can we be sure what is real and what is not?
With over 20 years of experience in the real world, doing real photos with real cameras and real lenses, I wanted to see what it was all about, and came to some initial conclusions after used the best AI art generators to try to make realistic images.
There’s no doubt that the most well-known art generators – including Dall-E 2 (openAI), Stable Diffusion (Stability AI) and Discord (Midjourney) – have a lot of potential, but they are far from replacing human photographers. in many areas, including the ability to convey original and meaningful stories. The following reasons explain why, for the foreseeable future at least, I believe we humans can take comfort.
AI images are faulty
Avid AI art creators will say that reality is the least interesting thing to create with an AI art generator, and that’s a good point – they can be good, clean, and simple fun. But the use of AI to create realistic images, without its use being disclosed, and even worse to manipulate viewers, is concerning.
My experience using what is currently the best performing AI art generator for realistic output – Midjourney’s Discord web app – has been mixed. By refining my word prompts to “photorealistic, window light portrait, black and white, female, 85mm F1.2 lens”, etc., I have had frightening success in producing head portraits and shoulders. But Discord, and other generators, can also produce ridiculousness.
Add other prompts, like “studio portrait, guitarist” and you’ll be greeted with a disfigured, twisted depiction of a musician holding a guitar. Who knows how many fingers the subject will have; or if they will have a thumb; or if the guitar is held the right way; or if the image even looks more like a photo.
This example is just the tip of the AI iceberg. How about “A man walking on a beach with a dog”…wait, he walks on water, and why are there so many dogs? AI can’t, it seems, bring happiness either. Think of all those hyper-realistic AI portrait images that fooled people – how many are smiling? None.
Happiness is like the “The One with the Engagement Picture” episode of Friends where Chandler can’t smile — the teeth are jagged white blocks, and even closed-mouth smiles in AI-generated subjects look forced. It’s like AI can’t or won’t generate happiness.
In the overwhelming majority of cases, AI simply cannot render reality.
The creative process – there’s really no struggle with AI
Visual artists undertake a creative process that can take a lifetime to master; learning to use the tools of their trade (and I don’t believe AI art generators are among those tools), developing a visual style, understanding how to tell a story and – to coin a phrase by the photographer, famous Youtuber and author Sean Tucker – finding meaning in the making.
AI art generators follow word prompts to produce results in seconds, scratching out what’s already there. Generators don’t explore the world, meet people, or make real connections – and it’s those connections to the world around us that are key to inspiring creativity.
You could say that in some ways AI generators are beneficial – creative visions can be realized on screen wherever you are, with no travel costs and no carbon footprint. But we humans need to experiment and create for our own well-being.
We need to strive to create something truly beautiful and truly meaningful. AI art generators can’t give us what we really need as human beings, unless I guess you really dig the art of thinking about word prompts, which is a quite a different kind of creative process; and if your prompts get lost in translation, you’ll soon be hitting your keyboard in frustration as the generator produces its strange rendition of your “vision”.
Detail Constraints
Whether it’s camera megapixels or a canvas, the size of the artwork photographers and artists can produce completely dwarfs the tiny resolution of AI art generators. By default, most of the best AI generators output square images at 1024 x 1024 pixels – effectively a million pixel image. Even on our phones we can take digital photos at 10x to 200x resolution, while canvases can fill an entire wall.
With a bit of know-how, you can double this output and change the aspect ratio with Stable Diffusion (another AI art generator), while the beta image editor used by Dall-E 2 can use a feature called Outpainting to extend an original image to a higher resolution by adding additional peripheral detail, but each time you add to an image you have less control over what the generator produces.
These limited output sizes could be improved in the coming years; and they will have to be if AI generators are to be taken seriously as an alternative to stock libraries like Getty, which offer much higher quality output. For now, though, we’re in the early days of AI art generators, and you wouldn’t want to see their images on a screen much larger than a smartphone.
Be accurate and up to date
In addition to the difficulty of rendering many subjects and genres, AI art generators will not suffice for very specific needs. They do the general concepts quite well, but you won’t be able to generate images of a friend’s wedding, or convince real landmarks and locations, images that feature designer clothes, or go overboard. with your word prompts.
Ultimately, AI art generators can only create output by scraping and regurgitating the millions of images that already exist on the internet. So while we can take A and add B to create an “original” C, technically there is no originality. World events yet to come. Artistic visions yet to be realized. Stories yet to be told. AI art generators just can’t do it.
So while those who increasingly adopt this technology can benefit from a quick and inexpensive turnaround, there are myriad aspects of the visual arts, and even of our humanity, that AI will never be able to touch. There is still life in us, photographers, and artists, and in each of us.