Will a computer ever be as smart as a human? By many measures, computers already are many times smarter than us. And yet, a five year-old child can learn to tie his shoes in a few short lessons, while a robot? Still waiting.
Perhaps it's the wrong question. "As smart as" implies similar to, and the intelligence that computers, robots, and androids acquire may be quite different from ours.
On ZDNet, a recent story with the arresting title, "This is your brain on a microchip," says:
Despite the false starts, many high-profile neuroscientists and bioengineers gathered this week at IBM to talk about how and why cognitive computing research is finally bearing fruit. Scientists from around the world talked about projects ranging from digitally mapping the human brain to developing microcircuits that can repair brain damage.
James Albus, a senior fellow and founder of the Intelligent Systems Division of the National Institute of Standards and Technology, made the most convincing case for why the era of "engineering the mind" is here. "We are at a tipping point...analogous to where nuclear physics was in 1905. The technology is emerging to conduct definitive experiments. The neurosciences have developed a good idea of computation and representation of the brain."
At the Singularity Summit a few days ago, both Ray Kurzweil and Eliezer Yudkowsky made the point that progress measured by exponential advances may actually seem very slow for a long time. The early years of the graph will show a nearly flat line, and then all of a sudden -- boom!
Note that in terms of results (the vertical), halfway to the goal is not achieved until more than 90% of the time and effort (the horizontal) have been expended. Many real world examples of this have been offered; you can probably think of your own.
Kurzweil and Yudkowsky -- and perhaps the cognitive scientists quoted in the article above -- might say that progress toward truly thinking machines is getting close to the exponential take-off point. Similarly, we at CRN have warned that almost all of the most remarkable results that will lead to a molecular manufacturing breakthrough will occur within the final one or two percent of all the time spent on the problem. And what that means is that we may not have much warning.