I’d like this to be my last word on the subject of AI generated art. Otherwise I’ll just end up sounding like an embittered dinosaur roaring into the impending meteor strike, waving a fossilised gingko walking stick at passing carnivores…
All of what follows is personal to my experience as a visual artist since… well – since when I discovered there was such a thing (as opposed to just being a kid having fun drawing imagined pirate ships, dinosaurs and apollo-inspired space rockets).
I don’t think AI generated art can ever be the same thing I do, because fundamentally the end result isn’t the prime reason I do it. I was making oil paintings for years before I had even the slightest inkling of an idea that there was a living to be made from it. It might seem strange to an audience where the internet and its offshoots seem to have been ever present and all encompassing, but in 1976, the year I first picked up a tube of oil paint, there was no notion of an ‘art world’. To a kid in a small north Somerset village, art was a subject of books full of interesting pictures and stories of dead people – usually blokes. The most that you would connect it to was perhaps some notion of ‘history’ because it all seemed to have happened and been done years and years before, but as a kid you rarely followed that up because the pictures were too much of a distraction. Or perhaps I just had the intellectual depth of a puddle.
Anyway… why do I make art? Essentially because I have found some need to make it inside me is fed by the doing of it. Perhaps when I was a kid it was the positive attention it garnered because I could make a pencil drawing give the impression of near photographic realism; having previously been the subject of substantial bullying, I liked the new positive attention… so I continued drawing and painting to keep getting that feedback. But in the end, it seemed that it was the one thing that I could do that I had full control over and so I just continued and obsessed over it. Eventually I started to discover the associated histories and theories of art, but that didn’t change the reasons for making the art. In a clichéd manner, it became like a paraphrasing of the mountaineer claiming that they climbed for no other reason that the mountain ‘was there’.
But (to coin another cliché… “it’s a cliché because it’s true”), the making of the work became more important than the end result. Certainly, from the early 1980s I was aware that there was the potential of having a career as an artist, but despite my increasing frustration and desperation of perpetually banging on gallery doors, I never had a painting in a commercial gallery show until 1993. But I couldn’t stop making the work. If I didn’t have money for canvas, I painted on old bedsheets glued to board, when I didn’t have oil paint I used children’s gouache or poster paint, when I didn’t have drawing paper I just drew on newspaper; my need to make drawings and paintings was insistent and obsessive – and the end results were rarely kept beyond a few showings to friends and family. I did sell the odd one or two, but any that left my hands to another were usually given away.
Where is this going? Well, friends know (because they’ve heard it in conversation) that if I never sold another painting, I would not stop making paintings. I can’t stop.
The programmers of AI generated art software only look at the processes to make an image as a means to an end, and unfortunately (the beast of our contemporary culture) that end is the financial reward realised by the production of a finished product. There are some artists, and I don’t criticise them for their position, who see what they do as producing a product solely for the result of putting bread on their table. I know artists who have given up making art because it is not generating the riches they hoped for. Perhaps they’re just being more realistic than I have been. Perhaps I’m nuts.
So, the top and bottom of it is that AI generated art, however pretty its results, will not have that attraction for me. It’s too easy; there’s no ‘work’.
I think this could play out the same way for the broader potential audience too. AI generated art has an intrinsic weakness (this may be algorithmically addressable – I don’t know) and that is its current methodology of referring, however marginally, to human created work as a source material. Sometimes this is even out of the digital hands of the machine as prompts such as ‘in the style of artist X’ are included when generating images. Consequently, and similarly driven by the financial need to appeal to the broadest possible market, the work will inevitably start to become limited to that tight financial, rather than creative, aesthetic.
I don’t know if this is a valid comparison or not… but take the last couple of decades of CGI created visual lunacy thrown at us in cinemas. As each new development in simulating apparent reality in depicting absolutely fantastical scenarios further challenged our belief in what we were watching was not real, the playing field for future films was being increasingly marked out into zones, halves and goals that had to be met by competing studios. The creatives’ aspect was becoming secondary to the accountants’ who wanted the box office returns that a previous film rolled in. Accordingly, we’re now at a point in time where so many of these films, particularly from the culturally omnivorous monolith that is Disney, are increasingly becoming exercises in tick-box blockbuster generation. The films are becoming increasingly formulaic, and the audience is already getting bored.
I said that I’d like this to be my last word… but while writing this another thought has stuck its head above the parapet – but I’ll save it for another day…

back to writing menu