The Latent Space Maps series explores image creation using a generative AI model based on Stable Diffusion. In the text-to-image process, positive prompts define what the model should generate, while negative prompts specify what should be excluded. This duality resembles additive and subtractive methods in art.
Interestingly, when the same element is both added and subtracted, the result isn’t “nothing” but rather a “phantasized” image—one that hints at the underlying “spirit” or essence of the AI model.
In this work, both the positive and negative prompts were nothing else then the word “negative”.