One of my MUS2640 students asked which AI tool I had used to create the illustration on top of the textbook I have been developing for the course. The fact is, it isn’t AI-generated; it is a photo!
I took the photo holding a 360-degree camera on my head while visiting the San Francisco Museum of Modern Art (SFMOMA) a couple of years ago. I was standing inside Olafur Eliasson’s One-way colour tunnel, located on a bridge inside the gallery. It is a fascinating three-dimensional light sculpture that no photo could capture. However, the 360-degree photo gives an impression of what it looks like:
The fascinating thing about 360-degree photography is that it enables capturing a space, but it isn’t easy to view those images afterwards. Such a photo can be “flattened” in many ways, which will give a unique—often strange—look. This is the full version of the flattened image I used for the textbook:
One of my colleagues was recently interviewed about “AI images are not fake photographs”. There she also criticizes the idea that photos are “real” and AI images are “fake”. I am old enough to have worked with film in a dark room. Even back then, we used tricks to remove parts of a photo by gently moving an object over the relevant part in the image while exposing the photo paper. Such tricks have also been commonplace throughout the history of “Photoshop” (I use quotation marks here to emphasize that Adobe’s Photoshop is only one of many photo editors. My preference has for a long time been the free GIMP.)
It is scary to see how photo-realistic AI images can distort reality, but it is also interesting to see people begin to question photos.