The machine learning (~AI) is all about code and data. If you planning to create something really unique, you'll probably need to, well, work with code and data.
You'll probably should look at github and try to get inspiration from opensourced projects. If you found some AI-related artwork, try to google it's name with 'github'
Another approach would be to look at medium, it's usually have some post/tutorial.
The art is very broad topic and it's hard to cover it all. I try to provide examples of different resources that could help you to start.
Google Research made some examples, where you run code in colab without installing anything locally. It helps you to try basic stuff.
There is tensorflow.js, that allows to launch code at browser. You can look at demos with code
There is intro course for visual artist on kadenze.
Here it is some more links on different courses.
DCGAN - very popular image generation library.
TouchDesigner, very popular tool among visual artists, provides python interface. For example, here is usage of aforementioned DCGAN with it
Also, there is creative section on NIPS(one of the biggest ML conferences) http://nips4creativity.com/. Some of the work provides detailed explanation, like this one
I know, starting to code could be intimidating, but many github repos provides some description about how to launch them and you don't even have to know how they work exactly to use them. Also, it would help to have a person with programming experience, who could help you to deal with code.