While working on my project to create abstract depictions of human pathogenic bacteria (see here and here), some of my prompts to Stable Diffusion didn't produce an image in the style that I wanted. But they still produced something interesting. For most of them, I have no idea what part of the prompt the AI latched onto or what similar word the AI went with. There were, however, a few themes that came up often. For example, if a bacterial genus named ended in –ella, the images often featured a woman. I guess it has figured out that –ella is a feminine diminutive. It also drew a lot of flowers, birds, and insects. My guess is that the training data set had a lot of flowers, birds, and insects captioned with their scientific names.
This was supposed to be Pantoea.
This was supposed to be Parabacteroides.
This was supposed to be Paraburkholderia.
This was supposed to be Paracoccus.
This was supposed to be Parvimonas.
This was supposed to be Pasteurella.
This was supposed to be Peptoniphilus.
This was supposed to be Photorhabdus. There seems to be an element of light in all of these images, which is what photo– means.
This was supposed to be Plesiomonas.
This was supposed to be Pluralibacter.
This was supposed to be Porphyromonas.
This was supposed to be Prevotella.
This was supposed to be Proteus.
This was supposed to be Providencia. This has to be Providence, Rhode Island.
This was supposed to be Psycrhobacter. I think the AI must have understood psycho– (mind or soul) instead of psychro– (cold).
This was supposed to be Rahnella.
This was supposed to be Rahlstonia.
This was supposed to be Raoultella.
This was supposed to be Rhizobium.
This was supposed to be Rickettsia.
This was supposed to be Riemerella.
This was supposed to be Roseobacter. Surprisingly, it only came up with a rose once.
This was supposed to be Roseomonas. This time it came up with roses every time.
This was supposed to be Rothia.
This was supposed to be Salmonella.
This was supposed to be Serratia.
This was supposed to be Shewanella.
This was supposed to be Slackia.
This was supposed to be Sneathia. The AI appears to have had a particular kind of woman in mind, but I'm not sure what it is.
This was supposed to be Stenotrophomonas. It looks like the AI was trying to paint beer steins.
This was supposed to be Sutterella.
This was supposed to be Tatumella.
This was supposed to be Terrisporobacter. At least three of these are weird motocross hybrids. I'm guessing that came from terri–, which means "earth".
This was supposed to be Treponema.
This was supposed to be Tropheryma.
This was supposed to be Tsukamurella. At least to some degree, the AI detected the Japanese origins of this name (after the Japanese microbiologist, Michio Tsukamura).
This was supposed to be Turicella.
This was supposed to be Veillonella.
This was supposed to be Vibrio.
This was supposed to be Weeksella.
This was supposed to be Weissella.
This was supposed to be Wohlfahrtiimonas.
This was supposed to be Wolbachia.
This was supposed to be Xanthomonas.
This was supposed to be Xylanimonas.
This was supposed to be Yersinia.
This was supposed to be Yokenella.
These illustrations were drawn using Stable Diffusion 2.1.
