Superintelligent machines have long been a trope in science fiction. Recent advances in AI have made them a topic for nonfictiondebate, and even planning. And that makes sense. Although the Singularity is not imminentyou can go ahead and buy that economy-sizecontainerof yogurtit seems to me almost certain that machine intelligence will surpass ourseventually, and quite possibly within our lifetimes. <\/p>\n
Arguments to the contrary dont seem convincing. Kevin Kellys recent essay in Backchannel is a good example. Hissubtitle, The AI Cargo Cult: The Myth of a Superhuman AIimplies that AI of superhuman intelligence will not occur. His argument centers on five myths: <\/p>\n
He rebuts these myths with five heresies : <\/p>\n
This is all fine, but notice thateven if all five myths are false, and all five heresies are true, superintelligence could still exist. For example, superintelligence need not be like our own or human or without limitit only needs to outperform us. <\/p>\n
The most interestingitem on Kellys lists is heresy #1, that intelligence is not a single dimension, so smarter than humans is a meaningless concept. This is really two claims, so lets consider them one at a time. <\/p>\n
First, is intelligence a single dimension, or are there different aspects or skills involved in intelligence? This is an old debate in human psychology, on which I dont have an informed opinion. Butwhatever the nature and mechanisms of human intelligence might be, we shouldnt assume that machine intelligence will be the same. <\/p>\n
So far, AI practice has mostly treated intelligence as multi-dimensional, building distinct solutions to different cognitive challenges. Perhaps this is fundamental, and machine intelligence will always be a bundle of different capabilities. Or perhaps there will be a future unification of some sort, to create a single capability that can outperform people onall or nearly all cognitive tasks. At this point itseems like an open question whether machine intelligence is inherently multi-dimensional. <\/p>\n
The second part of Kellys claim is that, assuming intelligence is multi-dimensional, smarter than humans is a meaningless concept. This, to put it bluntly, is not correct. <\/p>\n
To see why, consider that playing center field in baseball requires multi-dimensional skills: running, throwing, distinguishing balls from strikes, hitting accurately, hitting with power, and so on. Yet every single major league center fielder is vastlybetter than I am at playing center field, because they dominate me by far in every one of the component skills. <\/p>\n
Like playing center field, intelligence may be multi-dimensional, and yet one entity can be more intelligent than another by being superior inevery dimension. <\/p>\n
What this suggests about the future of machine intelligence is that we may live for quite a while in a state where machines are better than us at some aspects of intelligence and we are better than them at others. Indeed, that is the case now, and has been for years. <\/p>\n
If machine intelligence remains multi-dimensional, then machines will surpassour intelligence not at a single point in time, but gradually, and in more and more dimensions of intelligence. <\/p>\n
<\/p>\n
Read more here: <\/p>\n
Multiple Intelligences, and Superintelligence - Freedom to Tinker<\/a><\/p>\n","protected":false},"excerpt":{"rendered":" Superintelligent machines have long been a trope in science fiction. Continue reading