Screen Recording 2024-09-24 at 5.36.07 PM.mov
https://editor.p5js.org/yg2348/sketches/1FMzCSIDw
I found Ellen Nickles' research to be an engaging read that really got me thinking more deeply about data collection and how it's applied in machine learning models. As I reflected on it, I realized I'd love to know more about the diversity of datasets used for training these models. It struck me that adding information about a model's limitations to its biography would be really helpful - like explaining how biases in training data can affect its output. I also think it would be great to include a section on the ethical sourcing of data and how they protect the rights of people who contribute to it. That kind of transparency would make the biography feel much more comprehensive to me.
Knowing where a model and its data come from helps me create better art. It helps me make smart choices about using the model's results and spot possible problems or gaps in what the model understands. This lets me address these issues in my work. Understanding the model's background also makes me think about the ethical side of my art, helping me avoid accidentally promoting harmful ideas. It pushes me to be more creative, finding new ways to work around the model's limits or show off what it does well. In the end, this knowledge helps me make art that's not just cool technology-wise, but also thoughtful and meaningful. My work can be advanced, socially aware, and interesting to think about.