One of the advantages of new tech is almost always it allows you to do things never before possible. The applications that nanorobotics could open up are still on the drawing board. Until the technology matures, there is no telling where this will take us.
One of the tin foil hat groups will tell you of the fear of the
grey goo. Basically technology gone berserk and without control.
One the other side, medical techniques such as have always been considered impossible because of that lack. Remember the movie,
Fantastic Voyage? Instead of reducing an entire submarine crew along with the machine, what say we just do away with the crew and the machine. Instead you use nanotech to do the job. Or say cancer, where you have a vaccine that is only used if it is needed and exactly where it is needed as opposed to sending it everywhere through the body just to get those few offending cells. Many of the complications resulting in toxic type materials used in such vaccines would no longer be needed to be filtered out by the kidneys, removing putting those organs at risk.
Another foreseeable possible, is something akin to a
Von Neumann machice. When you send a population to Mars, they don't have to bring any construction equipment with them, because when they arrive, the things they would have build for living quarters, machine shops, excavators, and the like have already been built and are waiting on them to just move in. Along with a store of water and oxygen.
The idea here isn't to send the equipment ahead of the arrival of the colonists. It's to send a very small package of self replicating bots in nanosize. At first the bots make copies of themselves which in turn make more copies, till there's literally a factory of nanobots. Then they start making specialized bots that do just one thing, and then more that do something else. At nano size or smaller, they could mine the planet without having a mine. Creating machines able to break down molecular compounds into elements to reuse to create other compounds needed for such construction. Some where in all this would be needed a master program or plan on what to build when, how much is enough, when to schedule the altering of replication into specialized nanobots for certain tasks and when to do it. But theoretically, what you could pack into a 5 gallon bucket would be enough to start it all, given time.