Well, a logic gate doesn’t fundamentally have to amplify… if the control current exceeds the output, it isn’t amplifying but fill performs logic. I am too lazy to look myself, but did they demonstrate amplification? If not, I think it’s doable.
Couldn’t you build an amplifier by using a thin wire that heats up a larger wire? If you size the large wire to minimize self heating, then a small current would cause the thin wire to act as a heater, switching the large current.
For what it’s worth, it’s the same with Prusa. The only work I do on my printer is glue stick on the PEI sheet when printing otherwise-incompatible materials.
My quick and dirty math based on some captions of the figures from the paper suggest it’s unlikely they’re getting amplification for now, because it seems like the even the “low” resistance state is quite resistive. But I still suspect it can be done, and they do characterize their structures as “active” - thanks!