Will nanotechnology be viable within the next fifty years?

I think it’s interesting to compare the nanotech researchers of today with the artificial intelligence/robotics researchers of the 1950s. Some of the parallels are:

Nature gives an existence proof. We know it is physically possible to build robots which can walk, see, hear, assemble a bicycle, write a poem, etc., because humans and in some cases other animals can do those things. The robot can be built, we just have to be smart enough to figure out how. Similarly, we know it is possible to have tiny little machines working at a molecular level, because living cells, including bacteria, and their components, are just such machines.

At the same time, we’re a little vague on just how Nature does what it does. For both robots and replicators, opinion is divided on whether we should study the natural example and try to copy it very closely, or whether we should start with a more blank-slate approach and possibly use a very un-Natural design.

Nature can be improved on. If you’re smart enough to build a robot as good as a human, it seems likely you could improve that design a bit and build a superhuman robot. Likewise, a nanoreplicator could possibly outperform biological replicators, at least in some ways. There's probably no physical law against it.

Hubris. Artificial intelligence pioneers of the 1950s were quite confident that in a decade or two they would have computers that were as smart as humans and probably smarter. And then they thought that another decade would do it. And then another. Likewise, nanotech enthusiasts today are quite sure that they will have self-replicating nanomachines in a couple of decades---and they’re going to be way better than those dumb old bacteria, which are just the accidental product of dumb old evolution. The nanomachines are going to be “diamondoid.” (At the same time, like the AI researchers before them, the nanotech guys have critics saying, “You'll never do it,” and perhaps additional critics saying, “Your evil creations should be outlawed before you destroy the world.”)

What will come of it all? Reviewing the last fifty years of progress in robotics, it seems we have fallen drastically short of the early hopes. Imitating and/or surpassing the intelligence of natural creatures has turned out to be much harder than we anticipated. However, it has not been a complete failure either. We have robots of some meager intelligence working in factories, we have chess-playing machines that can beat the world champ—which the critics said would never happen, and so on. If you wish to be a little generous and count every computerized device as a “robot,” then we are surrounded by robots and they have revolutionized the world. Meanwhile, we'll keep working on the C3P0/R2D2 thing.

I have a hunch—it's just a hunch, by no means a logical necessity—that fifty years from now, the story in nanotech will be similar. The biggest dreams and fears (“you'll have this box, and it’ll make whatever you ask it to make”; “there’ll be this goo that turns everything it touches into more goo”) will not be realized. On the other hand, there will be more modest successes. And certainly, we will be surrounded by smart materials and things, which you could call “nanotech” if you wanted to, although the people who invent them may not think of themselves as pioneering Drexler-style “nanotech.”

How much will nanotech manufacturing revolutionize affairs? Potatoes and other crops are made (grown) via “natural nanotech,” which is why potatoes are cheap relative to conventional manufactured goods. But potatoes, though cheap, are not free. You still need materials (soil, fertilizer, water, sunlight) and some management (plowing, planting, watering, spraying with insecticide, harvesting), not to mention storage and distribution to consumers.