For the past few days, I¡¯ve been playing around with an app developed by the San Francisco company OpenAI that turns text descriptions into hyper-realistic images.
OpenAI invited me to test DALL-E 2 (the name is a play on Pixar¡¯s WALL-E and artist Salvador Dal¨ª) during its beta period, and I quickly got obsessed. I spent hours thinking up weird, funny and abstract prompts to feed the AI ¡ª ¡°a 3D rendering of a suburban home shaped like a croissant,¡± ¡°an 1850s daguerreotype portrait of Kermit the Frog,¡± ¡°a charcoal sketch of two penguins drinking wine in a Parisian bistro.¡± Within seconds, DALL-E 2 would spit out a handful of images depicting my request ¡ª often with jaw-dropping realism.
What¡¯s impressive about DALL-E 2 isn¡¯t just the art it generates. It¡¯s how it generates art. These aren¡¯t composites made out of existing internet images ¡ª they¡¯re wholly new creations made through a complex AI process known as ¡°diffusion,¡± which starts with a random series of pixels and refines it repeatedly until it matches a given text description. And it¡¯s improving quickly ¡ª DALL-E 2¡¯s images are four times as detailed as the images generated by the original DALL-E, which was introduced only last year.
With your current subscription plan you can comment on stories. However, before writing your first comment, please create a display name in the Profile section of your subscriber account page.