I am going to finally create a human cyborg with super-intelligence to beat an AI.

It's mid 2050's and my company Neuroconnect has produced a nanotechnology BCI (brain computer interface) capable of growing into brain tissue and connecting to every synapse.

What worked already early this century: the brain can learn about artificial limbs and eyes because it perceives the external capabilities with feedback.

But now, even the newest working chips still operate with 0's, 1's and logical gates. What should we do to tell the brain it can finally use a petaflop-on-a-chip computer, or connect to an exaflop smartphone? Would the brain "feel" a qubit interface maybe?

  • 1
    $\begingroup$ @L.Dutch: From the context, BCI= Brain Computer Interface. But the rest of the question isn't clear. Most of the answers would be: yes, if you designed it that way. $\endgroup$ – nzaman Aug 14 '19 at 8:14
  • 1
    $\begingroup$ That's not what you seem to be asking in the question. $\endgroup$ – L.Dutch - Reinstate Monica Aug 14 '19 at 8:30
  • 2
    $\begingroup$ It's fiction, so you decide how the interface works and how it feels. $\endgroup$ – Spoki0 - Reinstate Monica Aug 14 '19 at 8:38
  • 3
    $\begingroup$ All: you seem to be asking a reality check,but a reality check is "is this realistic given these constrain?", while your question seems to be "how do I do technobabble technobabble?", with the technobabble which is just technobabble $\endgroup$ – L.Dutch - Reinstate Monica Aug 14 '19 at 9:03
  • 1
    $\begingroup$ I am not talking about the BCI. I am talking about the last period of your post. $\endgroup$ – L.Dutch - Reinstate Monica Aug 14 '19 at 9:10

I can't understand how should be possible that a brain can integrate with logical gates.


But now, even newest working chips still operate with 0 and 1 and logical gates

So do neurons. You might think your wetware is analogue, but nerves don't actually work that way.

The magic phrase you need to know is action potential, possibly along with voltage-gated ion channel. Nerve signals are in fact trains of pulses, with reasonably well-defined on and off states. They're not nice square waves like digital logic would like to be, and nor do they use nice user-friendly communication protocols.

A wetware-hardware interface will mostly consist of a whole load of signal analysis and signal generation logic to interpret the messy organic stuff you work with and translate it into something machine-friendly and vice-versa. There's been a fair bit of work done on this sort of thing, especially with regards to artificial vision. Obviously we aren't in the realms of direct neural interfaces yet, but you've got to start somewhere.

Would the brain "feel" a qubit interface maybe?

I think trying to get a quantum-coherent communications channel from a computer into a brain is going to be harder than making an actual functional cyborg. There's some hint that brains do make use of interesting and (comparatively) long-lived quantum states internally, but they are unlikely to propagate far because human bodies are a hot mess and coherency is fragile. Stick with digital stuff. It'll work fine.

What it an implant would actually feel like would obviously depend on where you hooked it up. Remember that your brain doesn't have vacant expansion slots you can just plug new toys in to... every bit is used for something. If you're replacing something (eg. adding an artificial eye or limb) then you hook it up to the appropriate bits of the visual or motor cortices and hopefully it will feel and work just like the original squidgy version. If you're wedging in new functionality, then it'll feel like whatever the old bit of functionality did that you're overriding.

  • $\begingroup$ Thanks for your answer @Starfish Prime! the part with the override I can agree: so if you would "upgrade" the frontal lobe (logical thinking, reasoning) you would just effectively show better results in IQ tests and things "become clearer" to you (like in the movie Limitless)? Not sure whether our brain really has "XOR" gates etc. / neither if it really quantum perception capable. $\endgroup$ – J. Doe Aug 14 '19 at 9:15
  • $\begingroup$ But this is actually the answer: "of a whole load of signal analysis and signal generation logic to interpret the messy organic stuff you work with and translate it into something machine-friendly and vice-versa." Maybe this way we can fake that the brain has more capabilities to even further grow, extend and interconnect it already does; having even more and faster synapses might be good enough. $\endgroup$ – J. Doe Aug 14 '19 at 9:18
  • 1
    $\begingroup$ see also: quora.com/… $\endgroup$ – J. Doe Aug 14 '19 at 9:21
  • 2
    $\begingroup$ @J.Doe our ability to upgrade concrete things (sensory and motor things) seems like a relatively straightfoward thing, but upgrading abstract things like your ability for analytical thought is gonna have to remain scifi for now because we don't really understand how those things even work in the first place. $\endgroup$ – Starfish Prime Aug 14 '19 at 9:23
  • $\begingroup$ @StarfishPrime Even without understanding how it works, you might be able to increase functionality by hooking it up to additional “digital grey matter” by connecting your BCI to a computer that’s simulating regular brain matter. It’d be functionally like getting bits of additional brain transplanted into your head. $\endgroup$ – nick012000 Aug 15 '19 at 1:50

You should do some research on Artificial Neural Networks. Computer scientists have long been trying to emulate an actual brain using electronic circuits, with varying degrees of success. Nowadays those artificial neural networks power applications like Apple's Siri, Google's voice recognition and text-to-voice features and IBM's Watson.

A human brain should integrate easily with some kinds of ANN - it's like plugging in extra neurons. And if you could merge IBM's Watson with a human mind, we could probably make the world a better place since we would make lawyers obsolete.

  • $\begingroup$ Artificial Neural Networks are a category of computing architectures. The only link between ANNs and brains is that the original inspiration for ANNs came from the structure of the association cortex as described in popular magazines in the 1950s. It is neither easier nor harder to interconnect a brain to an ANN than to any other kind of computing device. $\endgroup$ – AlexP Aug 21 '19 at 15:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.