I think the proof is seeing how good diffusion models have gotten for making images. They're not perfect but they're leaps and bounds over what we had just a ye

I think the proof is seeing how good diffusion models have gotten for making images. They're not perfect but they're leaps and bounds over what we had just a year and a half ago. Many of these problems seem to have been unexploited simply on basis of nobody throwing enough gpu clusters at it yet.