Over the last two weeks, Google and OpenAI casually dropped a bombshell: multimodal image generation. Now, the same system that writes like Shakespeare after six espresso shots can also think visually.