Raw LLM Responses

You are being obtuse for no apparent reason. For you every real artist should then have to programme their own software from scratch to work. If you can't tell the difference between Art Direction and craftsmanship I don't know what to tell you. In twenty years if someone is painting something in the street it's probably going to be more admired than Photoshop popping an image for u as it's the same today. With today's AI you need to know a little bit of composition, lighting, etc. The whole point of this system is that the layman's layman can eventually use it. The same way that Copilot final form is not a coding assistant but agents that have complete context and sufficient as hoc knowledge to supplant the software engineers. An artist have a vision and the skill set they have to take it to fruition is the limiting factor. They do know see themselves as users of software it's a medium. Someone requesting Art from a Diffusion Model has a limitation of: 1. Is the art style of the that I want my image in the training set 2. Are there enough references of set objects in the training set All foreign factors that has nothing to do with the internal skill Most artist id they want and image they could painted and take a photo and digitalized as is. If you take a away the tool they can still do the artwork? Technically you could do the same with sculptures and some photo to 3D objects. The vision and technique is from the artist. When you are prompting you always have to take what's given or rejected until your preference is meet. If you cannot tell the difference in those two processes I don't know what to tell you man.