There's a news story replicating on the web right now about a "Functioning Synapse Created Using Carbon Nanotubes," for instance here and here. However, it's not quite as good as it sounds (yet) because it's just a SPICE (simulation) circuit.
Both of those example news stories include this partially relevant image with the caption "This image shows nanotubes used in synthetic synapse and apparatus used to create them."
It's a somewhat misleading way of saying that this is just a component of the synapse circuit. It's not the actual circuit. That picture is of a single transistor (a field effect transistor [FET]). Carbon nanotube FETs are not new. But it is a clue that the simulation uses transistors. And the circuit has not actually been constructed, so the "apparatus" photo there is kind of silly. It just gives the false impression that a synapse model was actually built physically with analog components.
So here is the image that would make more sense to go along with the new articles, which is a diagram of the actual circuit in question. This is from the original paper [1].
This may look a bit underwhelming. This is an electrical circuit schematic that in turn depends on certain SPICE models of carbon nanotube FETs (which have apparently been available since 2006). So in other words, this circuit is a particular model of a synapse being simulated with a simple circuit.
I don't know enough of this field to evaluate whether this is actually exciting or not--is this circuit good enough be used as a simulation (or as they suggest, to construct an artificial cortex) instead of more complicated and/or computer-based models?
References
[1] A Biomimetic Carbon Nanotube Synapse Circuit
Both of those example news stories include this partially relevant image with the caption "This image shows nanotubes used in synthetic synapse and apparatus used to create them."
It's a somewhat misleading way of saying that this is just a component of the synapse circuit. It's not the actual circuit. That picture is of a single transistor (a field effect transistor [FET]). Carbon nanotube FETs are not new. But it is a clue that the simulation uses transistors. And the circuit has not actually been constructed, so the "apparatus" photo there is kind of silly. It just gives the false impression that a synapse model was actually built physically with analog components.
So here is the image that would make more sense to go along with the new articles, which is a diagram of the actual circuit in question. This is from the original paper [1].
This may look a bit underwhelming. This is an electrical circuit schematic that in turn depends on certain SPICE models of carbon nanotube FETs (which have apparently been available since 2006). So in other words, this circuit is a particular model of a synapse being simulated with a simple circuit.
I don't know enough of this field to evaluate whether this is actually exciting or not--is this circuit good enough be used as a simulation (or as they suggest, to construct an artificial cortex) instead of more complicated and/or computer-based models?
References
[1] A Biomimetic Carbon Nanotube Synapse Circuit
A functioning neuron already! And you thought every neuron needs like a gazillion of synapses! Ha - not at USC engineering department they don't. And with the same scaling, since they have a neuron now, they have a brain! 2040? Pffft, already finished Ray, that didn't take so long now, did it.
OK, seriously, given the many diverse circuit models of synapses and neurons and neural networks for many years now without however people knowing yet what the full functions of a natural synapse or neuron actually are, to write that there is a functioning artificial system (even if they had the model physically working in the lab) would still be "Overhype".
With neurons, they have no more than models that reproduce the average firing rate. Real neurons can fire their spikes at precise times (first wave model) and nobody knows how they do that except for the obvious suggestion that the vast dendritic tree is some sort of calculating network. Real synapses and what they can do while we learn and all that is also just about being discovered.
So, just for claiming a functioning anything without knowing the function, you are already absolutely right: Overhype!