Unless you never use your phone or internet, you must have heard of people talking about A.I. Art — artificial intelligence that creates images or if you can say ‘art’, from textual descriptions. Here is a phrase that was entered into the software Midjourney: ‘Simpsons in the style of Tim Burton’ and this was the result:
Even though the possibilities are truly endless, when some people try to dig deeper into AI-generated art, they find it somehow frustrating, why? Because the AI is not always accurately capturing what you want to show – meaning when you type in the same phrase four times, the AI can interpret it differently and generate four (or even more if you want) versions of the ‘art’. It could be a happy accident, but is it always the case?
Because it is so difficult to control what the AI is ‘thinking’, who is the actual artist? The person who is actually typing the ideas into the software? Or the AI who takes ideas and briefs, and creates something we have never seen before? To me it looks like the relationship between art director and designers in design studios – the art director gives out briefs and direction and the designers do the work.
And if AI-generated images are now, obviously video will come soon. What will AI-generated video look like? Runway just released a text prompt AI video generator and if it is anything like this, it will be a game-changer.
Make any idea real. Just write it.
Text to video, coming soon to Runway.
— Runway (@runwayml) September 9, 2022
It’s obvious that AI-generated art is a genie that is not going to be put back in the bottle — it’s here. While some artists find this extremely scary or unsettling, can we find a way to make good use of it and maybe one day, everyone can be an artist?