AI Generative Art Based On Emotions

The following is an experiment on testing how four of the popular AI Art generative tools interpret emotions in the purest sense. Meaning just one word per prompt, no descriptive text, no mention of style, no parameters, everything at its default setting.
The results shown below are the first impressions, and for the two tools that start with a variation of four, I picked the first image of the four. No re-rolls and no tweaking or editing.
The four generative tools I picked for this experiment are DALL-E 2, Stable Diffusion, Midjourney, and Disco Diffusion. Note that the name of Stable Diffusion’s generative tool is now called DreamStudio.
Five emotional words The first interpretive image/art per each AI tool
Fear

Joy

Anger

Love

Sadness

Final Thoughts
As you can see, DALL-E 2 leans towards photographic realism, almost a bit too safe and reminiscent of stock photography. Stable Diffusion is the most random and surprising in results. MidJourney favors surrealism and fantasy, and Disco Diffusion highlights objects in abstract compositions with high contrasted tones.
Related Articles: Five AI Stories Compared Using AI to Reimagine Marvel Superheroes in Cyperpunk The Case For AI Generative Art






