3 ms·

> can we not skip over the two middle steps in brain-computer-input-computer translation Well, if we're talking Ghost in the Shell (the most iconic case of rob

• hinkley 3y ago
  I've seen it elsewhere but it's also possible those other cases were an homage to GitS. I think there's still an airgap if the human brain is not feeding to or reading directly from the external computer. To be clear, we are both treating the prosthetic as 'me' in this exercise, and the system we are communicating with as Other, which we cannot implicitly trust. The brain-hand interface could act as an input device, then you can still use your motor cortex to tell the hand to type a SHIFT-a, and the interface interprets it directly as an A instead of the movement for A. You can airgap an output device and still use it. There are some precedents for that in avionics I'm told, which are the moral equivalent of optical isolators meant to keep power spikes from frying the more delicate parts of an electrical circuit. Which I agree is a hell of a lot safer than directly telling your brain you see a virtual terminal, and that now is a good time to have a seizure.