Seems like you cannot access any information source these days without encountering some prediction regarding the impact of AI on our lives. Like that lady over there. I tried to hide a bit in fiction and so picked up Dan Brown's 2017 offering Origin. Bad choice. Interesting novel, but all about how humanity will eventually evolve into some sort of human/AI hybrid. Yech.
First point. AI isn't one thing. AI is simply a new way of utilizing and distributing information. The latest step in a evolution that probably began with hand gestures - "dangerous beast over there." Then speech, writing, printing press, the telegraph, radio, television, and digital media - of which AI is the latest multi-faceted iteration. And it is that "multi-faceted" aspect of AI that makes it an error to attempt to discuss AI as "a thing." It is many things with varying impacts in each environment.
Point two. I make no claim for deep insight into the hard sciences. But from what I have read, this seems to be where AI makes its greatest potential impact by simply reducing the time it takes to perform the multiple iterations necessary to legitimize results in a variety of fields - medicine, physics, chemistry, etc. However, one must keep in mind that human beings must first pose the initial intellectual seeds, the questions, that trigger these various AI endeavors.
Point three. AI doesn't always get it right. Some would assert that AI gets it wrong a significant amount of the time, sometimes with harmful, even fatal results. It strikes me that the error of considering AI to be one thing, might fuel some of these concerns. But we need to keep in mind that all versions of AI spring - initially - from human programmers creating computer code. Fingers on keyboards typing very fast. Then code talking to code. There is a very real possibility of the oops factor creeping in.
Point four. And I'll admit to a personal bias here - there are some fields of human endeavor where AI has, if no place at all, then only a very small one. I addressed this assertion more fully in this previous post: https://schragwall.blogspot.com/2026/04/artificial-intelligence.html
But I wanted to do a little reprise here. My concern is with distancing the artistic product from the hand of the artist. I freely admit that my own 2D artistic output is compromised by my inability to draw. Took a couple of drawing lessons when I was 10 or 11, but never kept it up. [See previous post: https://schragwall.blogspot.com/2026/07/practice-practice-practice.html ] So I do compromise a bit by sometimes stealing an established piece of art and repurposing parts of it as a canvas for my "doodles." [I'm currently working a piece I cribbed from Hopper's Nighthawks. Coming your way once I figure out just how I turn it into my concept: Skylarks.]
Anyhow, I feel that many of the proposed uses of AI in the arts - music, painting, 3D sculpture, dance, theater - just feel like a bridge too far. To fire up ChatGPT and ask it to create "a forest glade beside a river in the style of the Hudson River School" and have it actually do a pretty good job, reeks to me of the hand of Sauron, the Dark Lord of Mordor, or perhaps more accurately, a painting devoid of any creative hand. And the same goes for other "artistic" products created by/with AI.
I do realize that my bias seems a bit antediluvian, but fortunately, AI's grandparent, Google, allowed me to find a rather posh, although a touch dated, bit of support:
"Importantly, our report highlights that the creativity involved in artmaking remains beyond the capabilities of current ML [Machine Learning - rls] technology."
Ploin, A., Eynon, R., Hjorth I. & Osborne, M.A. (2022). AI and the Arts: How Machine Learning is Changing Artistic Work. Report from the Creative Algorithmic Intelligence Research Project. Oxford Internet Institute, University of Oxford, UK.
Yeah. Me too. Still.
