It would be great to see the movies of those Purkinje cells firing. It's gonna be a while before we have transparent skulls though... And as soon as 100 billion neurons are gonna fire, who is gonna solve this "big data" problem? That's a lot of data!
A lot of people still have blinders on and can't see past the imaging problems to the exascale analysis that's required to exploit these data, or just view the computational side as perfunctory and don't anticipate that it will likely end up being just as challenging, from a research and economic/organizational perspective, as the imaging and biology. I've been trying to get computation-focused research on massive graph and activity analysis in this domain -- especially exploiting multiscale, multimodal integration of disparate imaging techniques and models -- picked up for 4 years now without success and am finally packing up and moving on out of frustration with the myopia of funding and priorities.
Would you be interested in talking about this? I'm trying to get neuroimaging department to standardize methods for communicating data/models/findings and I'd love to talk to someone who's been through it.
Based on computing power alone you may be right, but human brains are idiosyncratic to say the least. While on one hand we are simple neural net algorithms, on the other we are dirty hacks woven together throughout evolution. Even given the former, if the latter lags behind AI will continue to feel distinctly unhuman.
I bet we'll produce a really interesting brain within the next ten years that doesn't seem human at all.
2. We now have detailed maps of parts of the human and mouse brains on the bio side as well as working AI-like systems (Siri, Watson, etc.) on the tech side. Who's to say the intersection of the two fields won't result in a human-like AI in 10 years?
3. The "law of accelerating returns" may work faster than anticipated.
