I tried making a grue with CLIP+FFT and got 1. a lot of construction cranes 2. some kind of bat thing.

https://imgur.com/a/UkPzcXZ

An interesting thing about CLIP is that when it doesn't know what something looks like, it instead generates pictures with the search text in them. That's why it confuses "an iPod" with "a piece of paper with iPod written on it".

I ran some Lovecraft through Story2Hallucination[1] which uses Big Sleep to make videos from text.

The results were quite something - https://m.imgur.com/tfWLsSR

[1] https://github.com/lots-of-things/Story2Hallucination