When one says "the majority is always wrong", what does one mean by this? Do we mean that the majority is incapable of making perfect decisions (supposing for the moment that it is possible for an external agent to objectively measure the excellence of decisions)? That with universal sufferage, a large portion of people will make their decision based on very badly imperfect information? Or some other criterion? And if they make decisions which are somewhat bad, how do you propose to put into power individuals who, on average, make better decisions?
The problem with autocracy, and the modes of gaining power in societies tending to autocracy, is that the only thing that it consistently selects for is aptitude in gaining and retaining power. The only cases where "gaining and retaining power" is beneficial for "a nation" — that is to say (albeit acknowledging my democratic and humanistic bias when I say this) the people living in that country — are when the general well-being and contentedness of the subjects play a significant role in the ability of the dictator to continue to govern. That is to say, in a dictatorship which is weak, and which must accomodate the masses, or fall to a coup d'état. A democratic form of government is the logical (and also the historical) progression of governments being forced to accomodate the wishes of the citizenry.
You could, of course, remark that the main selection pressure for career politicians in a democracy is also their ability to obtain and retain power. The question is then: what is necessary for them to do so? Different democratic systems have many failings, but the worst of them — suceptibility to lobbyists or commercial interests, being prone to personal conflicts of interest, and so forth — not only apply to officers in a dictatorhsip as well, but are the sort of problem that we would categorize as corruption, that is to say, a failing to attain the intended standards of the system.
A benign, honest, and strong dictatorship is conceivable of course: otherwise you wouldn't have thought to ask the question. However, concievability is not the same as probability, or reliability. Historically, we seem to find the conditions for a strong dictatorship to be benign rather... rare. Furthermore, "benign" does not entail "well-informed", any more than "popular" does: a ruler who is well-meaning, or even well-informed in some matters, can be badly informed in other matters.
For certain ethical and moral priorities, such as those espoused during the enlightenment in Europe — freedom of movement and settlement, freedom of speech, and so forth — a democratic form of government is the best general scheme that we have found to enact those priorities. This is not to say that these forms of government are perfect for achieving important objectives (such as, perhaps, ensuring the integrity of the environment), but we have no experience with any other form of government which is more reliable for meeting those objectives; and (to make a rather bland understatement) dictatorships have an extremely spotty record for achieving them.
