← Back to context

Comment by multjoy

12 days ago

The fact that it gives you these things means that humans would do it, because the training data includes exactly these things.

I'm fairly confident there's virtually no ethnically diverse nazis in diffusion models' training set.

It simply has a model of what ethnically diverse people look like, what nazi uniforms look like, and combined the two when asked.

The training data includes imagery that, when interpolated over a high dimensional manifold, results in these things.

That doesn't imply that they were in the training set, or even anything close to them.