There are SF stories about aliens exploiting literal human brains for alternative purposes than that of our being embodied agents with personalities performing acts in an environment. Heck, I suppose there are tales of our own future inventions exploiting our low-tech wetware for reasons that seem contradictory to their own available superior achievements / capacities.
But to overcome their solipsist-like isolation and develop into an intelligence like us, simulated brains would either need to be remotely linked to or embodied in a robot in a "real" environment, or receiving sensory input of being embedded in a virtual body / world -- or at least be supplied (somehow) with pseudo-memories of such events simply to have a base of conceptions and data to formulate dreams, imaginations, and sense of "personhood" from.
Whether or not a simulated brain could have experiences correlated to its sensory processing areas and to activity associated with introspective thoughts, is entering John Searle's Chinese Room territory, or the firestorms of debate ignited by functionalism versus his biological naturalism. That is, if anyone can figure out if what Searle means by "intentionality in human beings (and animals)" has anything at all to do with Chalmers' hard problem. Otherwise, if it simply concerns understanding language via expressing meanings with yet more language, or "intelligence" as in performing sapient behaviors (with robot body) or outputting wise or sensible sentences, then the substrate of hardware underlying computers, simulated brains, etc, is certainly capable of those affairs.
