August 10, 2022

AI learns to create images from text descriptions by destroying data

AI learns to create images from text descriptions by destroying data
AI learns to create images from text descriptions by destroying dataAI learns to create images from text descriptions by destroying data

A fresh approach to generating images based on text descriptions with AI, called a diffusion model, effectively un-destroys new images into existence

Technology 18 January 2022

Images created from text descriptions by GLIDE, a new AI

OpenAI

Early last year, artificial intelligence company OpenAI unveiled software with the surprising ability to create accurate images from text captions – even obscure inventions such as “an armchair in the shape of an avocado”. The company has now released a new AI model that is smaller but capable of producing even better results.

Last year’s program – called DALL-E – was a large, 12-billion-parameter AI model that was trained on a huge set of images with associated captions. In recent years, most progress in AI has come from this sort of approach; training them …