It is obvious that many of the most powerful new technologies are likely to flow from biology, but one of the most game changing is likely to be neural control of devices. We are not far from being able to "jack in" to the Web. Why do I think so? Several new biological tools are converging to give us both an understanding and new capabilities at the neuronal level.
The first new tools are the means to image the internal working of living cells, including neurons. The second are a variety of tools for precisely mapping complex bio-molecular mechanics so that we can understand and manipulate neural functioning within the cell at a molecular level. And finally the functional MRI is already giving us a systemic understanding of neural behavior.
Over the next several decades there are likely to be other significant new biological tools that I have not foreseen that will only strengthen my argument. But the combination of these three are already likely to give us sufficient insight into how the brain works that we will be able to construct the means to reliably read the state of the brain and use that information to control external devices.
The first steps toward that are already far along in the pursuit of advanced prosthetic devices. The ability to give the seriously injured the ability to control a prosthetic arm is already a reality in the laboratory. And we have read out neural states that seem to express language. It is a few big steps from there to the ability to reliably control devices.
But if this does come to pass the most obvious applications will be the control of such physical devices as cars, trucks, fighter jets, drones, machine tools, etc. But the most interesting device to control will be the computer cursor/keyboard. It is not hard to imagine a piece of technology, say like a blue tooth ear-piece that would enable one to think of a message and send it. It will be a form of one way electronically mediated telepathy. Reading out a message to another person will be fairly easy in the sense that control of the keyboard, will give one the ability to transmit.
But we are much further from understanding the read-in process. We may be able to send fairly easily but we may never figure out how to receive directly into the brain. We may have to read the information as we do today even if it is a neurally expressed message. It might be some advanced form of interface, e.g. electronic contact lenses, but it is just a better computer screen. But neuro-transmitting by itself will be sufficiently game changing, and also will be a step along the way to a radically different world of computer mediated reality in every sense.