The first computer art tools were pixel editors, the apex of which was Deluxe Paint for the Amiga. Then Photoshop came in and changed the paradigm to image editing: old schoolers laughed at it because its pixel tools were so weak, but there was magic in changing the conceptual focus to the image and often editing it at a much higher resolution than the display. I’ve come to recognize generative AI art as the next paradigm change, away from drawing the image, towards specifying a scene through a combination of prompts, infills, and hand edits - whatever gets to the best result fastest. Fine Photoshop artists lament it (and shady IP practices warrant their respective criticisms) but ultimately it’s a new tool with a new paradigm that will likely become the predominant one in the future. And though the current tools are all about images, they’re rapidly evolving into video and 3D scenes, the ultimate version of which may be full 3D scene and animated shot synthesis and tweaking at professional levels of quality. For those born after the birth of the personal computer and the internet, it’ll be one’s first contact with a complete technological upheaval. The rapid burst of progress over the past 18 months in chat and art may be a bit misleading though, as they were preceded by 3 decades of directly applicable foundational research. Some areas of future AI application are lacking in these foundations, so we can expect progress to come in fits and starts. But the exponential upward trend is unmistakable.