Ethical issues with AI
When it comes to using AI to make and change art, there are multiple benefits and drawbacks that must be considered. AI has the ability to help us better understand art and can even give us a glimpse into the past by utilizing enhancement and coloring tools. However, these tools can distort and change the art in ways that can reduce its significance. According to Sonja Drimmer, “Adding color does not show things as they were but recreates what already is a recreation.” This is arguing that although adding color can make the image more modern and easy to interpret, it can change that interpretation from what the original image is trying to portray. It can cause issues as to whether it’s actually preserving the history of the images or reinterpreting and changing it. Another dilemma when it comes to this use of AI can undermine the meaning of art. Art represents the troubles, values, and creativity of a certain time period through expression, and according to Drimmer, “A computer algorithm cannot perform these functions.” AI changing photos can take away from the significance that comes with the creation of a piece of artwork. It undermines the original artist and can take away from the meaning.
MY EXPERIENCE WITH AI COLORIZER AND ENHANCER
When it comes to my own experience with AI enhancers, I had a very smooth and simple time. I pulled an image from the archives and was able to enhance it. Following this, I uploaded it to the colorizer, where it was able to add some simple color and didn’t seem to alter the image too much. I also used the distant viewing explorer to see what categories it would sort the image into, and this was the biggest issue. It was incredibly inaccurate with how it categorized the picture. It put some random labels on there, and I found it kind of funny how far off it was. The original image was zoomed out and blurry. The enhancement made it zoomed in and clear, and the colorizer seemingly brought it to life with how it added color to it. Overall, I think that these tools are useful in certain settings but need to be used carefully, and the original picture and creation must be kept in mind.
I have to wonder what the Distant Viewing Explorer AI was trained on as I also had a result that suggested “fire screen, fireguard” and I think a lot of other people did as well. I thought my results were wrong, but somehow yours are far worse. Rotisserie and axolotl are such random conclusions.
I also used the same quote from Sonja Drimmer. I really enjoyed her article, but I did disagree with some of her points towards the end. What did you think of the article?