After our last class session, I revisited my proposal and decided that I wanted to focus more on the negative aspects of AI. I reasoned that I have concerns about the technology taking away jobs and producing lazy/soulless work. Plus, Montserrat students and faculty are more likely to be receptive to the exhibition.
I also discovered crucial information about Adobe Firefly by conducting online research. I learned that Adobe never gave users uploading stock assets a sufficient opportunity to opt out of training the AI before its implementation. In addition, there’s a possibility that Firefly’s model was contaminated by images that were created on MidJourney. For context, MidJourney is an external AI model trained using copyrighted material all across the internet. These two factors create a massive problem for Adobe, who advertise that their AI is completely ethical and copyright-safe.
From here, I would experiment with creating AI typography. My process included removing pieces of different fonts and using generative fill to substitute the parts that were missing. This didn’t end well as the AI focused more on generating texture than comprehensible letterforms.
I realized that I needed to get a handle on how generative fill worked before I could take on such a feat. I proceeded to make a collage of amalgamated human features.
No comments:
Post a Comment