This post contains a lot of questions and very few answers, I hope you enjoy it regardless and it triggers some thoughts. Back in March, OpenAI demonstrated their newest model “GPT-4V” (the V is for Vision) on a dev stream: In this YouTube video at 16:19, Greg Brockman demonstrates turning a rough sketch of a website into a working prototype, almost instantly. In this example GPT-4V not only interpreted the image, but went on to write the HTML, CSS, and JavaScript for interactivity. It al...
