Re: Uploading with current technology

From: outlawpoet - (
Date: Sun Dec 08 2002 - 12:13:51 MST

--- "Ben Goertzel" <> wrote:
>Can anyone with more neuro expertise tell me: Is this guy correct as regards
>what is currently technologically plausible?

Insofar as I am aware, his approach fails in two areas. I haven't heard yet of an experiment that tricks Neurons into interacting with manufactured interfaces as-another-neuron. Most BCI(ish) devices we have now don't interpret native signals from the brain, they require adaptation on the part of the user(donor?). Also, His idea of slow interpolation of microsensors into the brain, it's not a new idea, it was featured in some nice science fiction about ten years ago in my now defunct favorite magazine OMNI, unfortunately, Insofar as I am aware, the capability to place such sensors in a fully invasive way is not yet available. The brain is an array of neurons in 3d at high density, and nobody I know of has solved how to place sensors within targeted and comprehensive ways.

That said, it's a nice thought, and a fun webpage. I currently intuit that he's right, and that you could upload somebody with current technology, but I'm not sure that's how it would work. In the area of neuronal simulation, there's a University of Utah project that gives full-axis control of a robotic arm (sans full motion hand), and it requires a top of the line laptop for processing power, just to interpret the captured nerve impulses. That for the equivalent of 6 channel radio control. So the guy forgets to mention the probably nonexistent massive supercomputer(cluster? 3d chip? who knows?) that would be required even in the early stages(localized impersonation)

Plus, I'm not sure that current neuron models are up to the simulation, but that's not my field.

I could be off-base here though, anybody onlist involved in Brain-Computer-Interface?

Justin Corwin

"They say you only live once. Well, if you lived I did, once is enough."
                            ~Frank Sinatra(immortality proponent)

This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:41 MDT