![[OLD SEMESTER] 15-104 • Introduction to Computing for Creative Practice](https://courses.ideate.cmu.edu/15-104/f2021/wp-content/uploads/2023/09/stop-banner.png)
One of the things I’ve found very interesting is the use of AI in generative art,
whether in the 2D or 3D sense. I recently came across DALL-E a neural network
that was programmed to be able to generate images from user-provided text
captions. These captions could be as arbitrary as possible, used to create 2D or
even 3D images.
Following my curiosity in similar computational programs, I found artist Matteo Rattini, who similarly trained a neural network to create images of contemporary sculptures based on the Instagram algorithm. The sculptures themselves were all very interesting, taking on forms that seemed organic or close to those seen in nature.
One of the most intriguing points of this program was how it related to
current society’s behavioral patterns and trends. Because sculptures were generated based on a user’s Instagram algorithm, what Rattini found was that while Instagram suggestions would initially be wide and diverse they would always eventually reduce down to a very narrow range of things, causing later sculptures to eventually all look the same. In many ways, I find this interesting, in that while creating sculptures the neural network is also subliminally reflecting the ways users interact with social media as well as patterns in current trends.
More about Rattini’s work: https://hyperallergic.com/667011/art-of-ai-matteo-rattini-eerily-real-computer-generated-sculptures/