New Video: IS AI JUST A TOOL?

This video combats the often-repeated argument that AI is just a tool when it comes to making AI art, and the real art is “the idea”. This same belief underpins a lot of contemporary conceptual art rhetoric and has its roots in the anti-art, anti-aesthetic, and anti-painting movement of the early 20th century, particularly in the jargon surrounding Marcel Duchamp. I argue that visual art is rife with ideas in visual language that are outside of linguistic thought, and the real meaning of paintings is wrought in the painting and the process of painting itself. Therefore, the ostensible negligible “tool” doing worker-bee errands is providing the real ideas that are expressed directly through the image.

I converted my last article into a video with a lot of extra, extemporaneous commentary, including about the AI art I used as examples.

Subsequent to making the video, I significantly updated the article, so as it turns out, the video and article each have content that the other does not. The reason I updated the article after the video is that I was going to redo the video based on some constructive criticism and pushback I received. Eventually, I decided to keep them both as they were.

The criticisms were concerning the “stream” aspect of the video, which is the impression that I did it all live in one setting. That’s not actually the case, and I’ll take it as a compliment if it seems that way. There are multiple takes, editing, and splicing together that worked better than I’d thought. Either way, some people weren’t overly fond of my delivery or that fact that I read from my article. I get it, though if I weren’t visibly reading from my article, I’d be doing it behind the scenes. I guess they don’t need to see me doing it.

I like getting feedback because it helps me tinker. I like to have some live elements to put a face on the videos; even if I’m about 50/50 on it, that works for other people. Since the onslaught of AI, where people are using ChatGPT to write their video text and using AI to narrate it, I may be moving in the right direction by introducing a fallible human. However, I need to keep it in the right proportion.

I believe this is now the future of art and commentary: AI will serve up a kind of perfection, but we will come to value the human touch even if the result is not as good in many respects. This is the case with auto-tuning people’s vocals and using drum machines that get every beat exactly on time. The results lack resonance. Something about the stain of a human operating within the boundaries of time and physical constraints gives something the stamp of authenticity. Further, I think there’s a great likelihood that the information AI gives us will be perfectly packaged and persuasive, but beneath that shiny exterior lurks a behind-the-scenes tinkering and controlling so that the AI tells us what is, as usual, in the best interests of the most powerful people and institutions.

As people begin to detect robotic content, the hunger for real human voices and visages will grow. Perhaps this video is a flawed experiment that nevertheless makes some solid points. Let’s see if I can tease out the strengths for the next video and incorporate them into my other techniques.

And if you are thinking that AI will be trained to fake being a mortal human with limitations and fallibilities, you are right. It’s going to be difficult to sniff out the imposter Cylons and Body Snatcher. All our old sci-fi dystopian books and movies are becoming reality. They were supposed to be warnings, and not How To manuals. We average mortals on the receiving end can learning from the great literature and film how to resist. We are now all becoming the John Connor, or the beleaguered crew of the Battlestar Galactica.

~ Ends