Grappling with AI and Art

When we started working with colorization, one of the first ways I thought about it was with landscape photos. Like it was pointed out in class today, AI colorization of nature are probably a lot more accurate because outside lighting and general tree/plant/sky color stays the same. In the photos that I colorized from the Carleton College Archive (see below), I think it did a better job than most of the interior photos, albeit the colors are still a little washed out and inaccurate like the top left and right corners of the photo.

One somewhat ethical thing I am worried about is that AI colorization might take away from the original meaning impact, and intent of the original photo. The first person that popped into my head was Ansel Adams, but even now, photographers take black and white photos to emphasize their message. I also think it invalidates an important profession, kind of how Sonja Dimmer in How AI is hijacking art history says that “these studies and projects also seem to promote the idea that computer scientists are more adept at historical research than art historians”. Just as “art historians study the ways in which art can offer insights into how people once saw the world”, I think that black and white photos offer different emotional insights than colored photos. Even more than that, I think that black and white photos marking dated photos is important for understanding history and context. If every photo was colored, I think we would lose an important part of history.

Out of curiosity, I put my favorite Ansel Adams photo in the AI colorizer and it just doesn’t have the same pensiveness and meaning once its colored.

Omeka Link: https://dgah.sites.carleton.edu/digtialobjects/admin/items/show/198