One ethical issue that stems from the usage of artificial intelligence in combination with image manipulation is that many things could me misrepresented thus leading to the spread of misinformation.
“The edited photos therefore do not give us absolute certainty about the actual colors.”
“From black and white to color”, 2024 Ghent University https://www.omeka.ugent.be/interieurdesign/s/plaatsdelict/page/kleurenfotos#
Once an image is manipulated and posted online, it is often times taken as the truth when really it could just be an interpretation of the image. That interpretation of the image could be a harmful one, especially if it changes a fundamental element of the image. In class we were shown several AI manipulations of an image of a women. The main thing that changed among the images was her skin tone. This could have serious implications if the same thing was to happen to a collection of many similar images. Not only is the woman in the picture misrepresented but if that information was shared as the truth to a large group of people, for example in an article or a study, it could be quite deceptive since the information portrayed in inaccurate.
“But this effort to “bring events back to life” routinely mistakes representations for reality. Adding color does not show things as they were but recreates what is already a recreation – a photograph – in our own image, now with computer science’s seal of approval.”
Drimmer, Sonja. “How AI is hijacking art history“, 1 November. 2021, The Conversation, https://theconversation.com/how-ai-is-hijacking-art-history-170691
Drimmer’s article explains how artwork is “discovered” by artificial intelligence when it had already been known to people, such as art historians. This supports the notion that “computer scientists are more adept at historical research than art historians” which is not necessarily true. Going back to an argument from past blog posts, the computer science world and the humanities world could greatly benefit from working with each other. But that does not mean one is better than the other; computer science in this case. This discredits the work and capabilities of people like art historians which could potentially misinform people about the field.
3 thoughts on “Blog – Week 4”
Leave a Reply
This site uses Akismet to reduce spam. Learn how your comment data is processed.
You highlighted the negative impact of AI coloring on various groups, a concern I share. Embracing AI, we tend to prioritize “improvement,” inadvertently neglecting the authenticity of the original and the groups involved. Your choice of quotes is particularly representative of this issue.
Hey Sunniva! I totally agree with you and I also had a similar reaction when we were shown the image of that woman in class. The misrepresentation of images throughout our history can be extremely damaging, and I think you brought up a really interesting point that it would be even more damaging if an entire collection of images was tampered with in this way. Great post!
I definitely agree with you about all the potential for misinformation through the misrepresentation of AI-colorized images. I feel like there have to be certain ways to avoid AI being biased in this way when making these images. I think your quotes get to the heart of the issue many have with this which is that the colorized images aren’t reality but just a recreation and a guess at what these scenes looked like.