[cleandreams]

I think Stable Diffusion is a both a threat and a tool for illustrators but will mostly be a tool for artists.

The reason is that AI intelligence is probability generated mimicry. By extension it mimics art but is not art. Think of the character arcs in Shakespeare, the narratives, the layers of meanings. This sort of thing cannot even be modeled as a problem in AI. The ability to define what is necessary to produce such layers of meaning does not exist.

I think AI is missing something fundamental about intelligence. What it is missing is the ability to construct systems of abstraction and have them relate via meaning (which is informed by history, context, experience).

Something new must emerge for AI to move past probability generated mimicry.

[AbrahamParangi]

I think if you dig into the latest research you would find that AI is much closer to what you’re imagining than you think.

One of the lessons from DALL-E and latent diffusion is that problems can often be decomposed into a “semantic” space and a “textural” space loosely corresponding to what we consider layers of abstraction.

[pyaamb]

Can Dall-E be extended to generate 3d models?

[ryoshu]

https://www.kaedim3d.com/ is working on 2D -> 3D.