[blueblimp]

This was definitely easier to follow.

Since they're building a special-purpose accelerator for a certain class of models, what I'd like to see is some evidence that those models can achieve competitive performance (once the hardware is mature). Namely, simulate these models on conventional hardware to determine how effective they are, then estimate what the cost would be to run the same model on Extropic's future hardware.

[Eliezer]

Ah, but running an experiment like that risks it returning an answer you don't like.