Luminar AI

There is a new kid in photography town. Not sure if I like him.

As a long time follower of Thomas Heaton on YouTube, I came across this video of him in which he expresses his feelings about AI (Artificial Intelligence in (landscape) photography. Referencing to this promotion video for Luminar AI, he asked for opinions on AI in photography. Well, here is mine.

Incredible things can be done with AI to pictures, and we’re only touching the very first products doing this. Imagine what AI can do to pictures in 5, 10 or 20 years. Looking in this era beyond 20 years on a technological perspective, is virtually impossible anyway.

What is a picture? What is photograph?

That is a very good question, albeit it – besides from the obvious characteristics – very hard to define. A (good) picture is an images that moves you in some way. That touches you. That makes you wonder. That inspires you. In relation to AI in photography, you could argue that any image that does this to you, has the desired outcome, despite its source. Is it a painting? A drawing? A photograph? An etch? … Or is it a CGI? (Computer generated Image). As long as it has the desired outcome, this is a good thing, and the source should not really matter. So, thumbs up for AI if it makes you produce more outcome that people like, no? Actually not. The question is, why would you use it? To generate more likes on your product on social media, so your ever increasing addiction to dopamine (dopamine is what your brain produces whenever you experience something good and makes you feel happy. Your brain gives you a shot for every like) keeps getting fed? Because, after all we live in an era where you have to ‘stand out’ and ‘be liked’ (preferably by people not giving a rats ass who you are or what you do, and you feel the same about them as long as they are large in number) for the stuff you put online. Next to that these AI tools ‘make everything look better’, and it will make this happen in seconds. Let me point out a few problems here…

  • What is ‘better’? By whose definition? The software developers? who are they to decide what is bad / good / better / best? And why are you willing to comply with their ideas of that – and even more sick: you’re willing to pay (the license fee) to comply with their views… Do you feel awkward already?
  • The product is praised for being able to ‘slim down’ people. To alter the way you look in such a way that you’ll be able to feed that dopamine addiction even more. But it is not you. The likes are not for you. They are for somebody that does not exist, but is ‘based on you’. So if you as a model have any integrity, self respect and pride, demand of anybody you work with, to not use these tools. Yes, there will be temptation. Resist.
  • Altering skies, landscapes, scenes, … So where is your added value? For what are you actually taking credit in the work that is produced…? Basically for taking a poor picture, but being able to pay for software that hides that fact that you are a poor photographer faking his/her way to the very much needed likes and picture sales. This sounds like a common thing to do in our every becoming more superficial society. We’re losing ourselves. Read books about that.

Should AI be forbidden? Banish it for all eternity?

No. You can do cool stuff with it. And it will only improve. But it should not be mixed with photography. It is / can be an artform in itself. And there is a lot of value in that. There is no problem in a new artform emerging being digital or whatever kind. But I feel there should be a clear distinction between these artforms. Very often, new technology is frowned upon. Let’s take photography as an example. When photography was invented, it was seen as a lame way for people too lazy / stupid / dumb / … to draw or paint, to represent reality. Over the years it developed as a separate, complete distinct artform, with – for most of us – no relationship whatsoever to drawing or painting any more.

My hope for the future

Nobody knows in which direction this will go, but I can tell you my hope for the future regarding AI in images. I hope it will be so good at what it does that it does not need to rely any more on photography in the first place. I hope it becomes that good that it will be able to be an artform in itself. I hope that people will start recognizing what they see is a CGI and appreciate it as such. And please, let there be a lot of people inspired to become ‘digital artists’ and use this technology to the fullest.

In the last 20 years, there was a rush on photography. DSLRs (and now mirrorless) suddenly became within everybody’s reach (read: budget), and because of it being digital, the workflow was accessible for everybody (putting a card into a slot in your computer is way easier than fiddling around with chemicals hoping to put that image on paper). These days, with high end smartphones and ditto cameras, even the DSLR and mirrorless products are challenged. The question is, what does the one pressing the shutter have in mind? Create a memory? Create a source for their dopamine addiction? Well, this is all certainly within range of smartphone cameras. Go for it. Use it. Enjoy it. Use AI. But that is a different goal than what I define as a ‘photographer’. And maybe – who knows – the rush on ‘everybody is a photographer’ will slow down, ad the rush will be on ‘everybody is a CGI creator’. And that is what I really hope. This will leave more room for people who want to create photographs to stand out in the art they love. It will be a smaller community, but a better one than it is now. It will make people appreciate photography again. As long, of course, DSLRs and mirrorless cameras are being produced, and spared from this AI interfering with the process. But I have high hopes. After all, even today, you can still buy rolls of film, and even after the death of Bob Ross, people do still paint…