The reason for the brain gap between humans and machines is the architecture of the machine hardware, operations system and application software is too limited. Unless, and until, machines can be designed to “think” like a human brain, they will never catch up. Right now, no computer, ounce for ounce, is a match for the power and speed of the human brain. Indeed, though there are some very advanced computational devices that can out perform a human brain on some narrow, specific task, no such device exists that can compare to the flexibility, analytical power and intuitive capacity of the weakest human brain, even mine.
Feel good about that? Feeling safe? Don’t.
The Defense Advanced Research Projects Agency (DARPA), an agency of your United States government, has been funding a development project by IBM, Stanford University and four other universities, to create a machine that can think like a human brain. In less than a year, the IBM project has succeeded in developing a system that simulates, and exceeds, the computational power of a feline. That’s right, there is now a computer that can out think Tabby, maybe Tabby and Mr. Boots combined.
Lest you think this is simple stuff, though I have frequently been out smarted by cats, this requires more than a billion “spiking neurons” and over a trillion “individual learning synapses.” I am not sure what either of those things is but I recognize that there is one hell of a lot of each.
According to an article in Cnet News, Dharmendra Modha, IBM’s Manger of Cognitive Computing Initiatives, the DARPA project should make it possible for the team to simulate an actual, functional, human brain within ten years. This will mean bridging the gap between the average cat cognition and the average human cognition level, a twenty-fold increase. This will be possible because this project, for the first time, is developing an accurate neuro-cartography of the human brain. In English, they are mapping the internal communications system of the human brain.
Believe you me, I can see the amazing potential benefit this new map could bring to all humanity. Last spring my daughter was in a terrible auto accident. Fortunately, she is on the way to full recovery but it was touch and go for a while. A major injury she sustained was a fractured skull that later required brain surgery to remove a hematoma. I can easily see how the type of information this project will, as a by-product, develop could make surgical brain repairs, as yet undreamed of, realities in my lifetime. Birth defects, damage due to injury, who knows what all, may soon have a medical resolution currently unimaginable.
On the other hand, as Dr. Mudho was reported to have said, this research will result in new computer architectures and will allow machines to deal with the “tsunami” of data that will be uncovered and processed in the near future. He used an accurate mapping of the world’s water and hydrology systems as one potential example. He used an accurate predictive model for the world’s financial systems as another.
Who can possibly say to what uses this new technology will be put? The only prediction I can make with absolute certainty is it will have us all hankering “for the good old days of 2010” before we know it. Well, that and, of course, the science fiction writers will have a field day.
It does make humanoid robots less unbelievable. Also, more conceivable now are new projection systems that make holographic displays commonplace. While commonplace holograms will make video conferences a real jazzy thing, they will also bring a whole new dimension to teenage sexting.
Clearly, we are going to need more internet spectrum than we have now. Maybe the robots can build it.
Prior to the tragedy at Fort Hood, this kind of technical breakthrough was the kind of thing that would make me all nervous about the government. Who knows what some future Dick Cheney might do with this kind of computational power? It doesn’t bear thinking of.
However, it may not matter as long as the machines don’t get too much smarter than we are. I mean, Major Hasan, the alleged Fort Hood shooter, was known to just about everybody. In hindsight, there seemed to be all kinds of warning signs that could have and should have been heeded. It wasn’t a lack of information; it was a lack of understanding.
I wonder when IBM will get around to working on understanding. Humans have been thinking for millions of years and nobody seems too concerned with understanding. At least, we seem no closer to that today than we were at the dawn of human history.