In Milano Luiss Hub where is a makers hub, Luca Stornaiuolo explained some basics of how do they train an AI for image processing and creative work and he also presented the AI that they are developing to create some arts and designs. He was speaking about the generative adversarial network (GAN), CycleGAN, Pix2Pix and more terms in artificial neural networks field, and I am not going to explain those words. The main problem was the talk was in Italian and I was not able to understand his explanation. Anyway, at least the slides where in English so I could get some points.
The main idea behind this program was enhancing and improving design experience by providing a tool for designers based on machine learning and neural networks. The tool was in its early stages and needs much more images to learn from and create its generated designs. But for now, it was able to do some interesting collages. You can draw on one side and the machine colors it and gives different materials and texture to it. It is nice to have some recommendations for what color or material should your design has.