“Most AI images look like shit. AI “artists” are quick to lecture me that generative tools are improving every day and what they spit out won’t always look this way – I think that’s beside the point. What makes AI imagery so lousy isn’t the technology itself, but the cliché and superficial creative ambitions of those who use it. A video of a cyber-punk jellyfish or a collie in sunglasses on a skateboard generated by Open AI’s new video-to-text model Sora aren’t bad because the animals in them look unrealistic; they’re bad because they’re mind-numbingly stupid. AI image generation is essentially a truncated exercise in taste; a product of knowing which inputs and keywords to feed the image-mashup machine, and the eye to identify which outputs contain any semblance of artistry. All that is to say: AI itself can’t generate good taste for you.”

—Elizabeth Goodspeed on the importance of taste – and how to acquire it (via It’s Nice That)