AI, The Human brain, and Real Trade Offs in Life.

It’s ironically interesting that if you’re not doing work that has tradeoffs, it’s probably not all that interesting.

I have a soft spot for following, but not particularly researching, artificial intelligence (AI) and neuroscience. Something I’ve spent significant time thinking about is whether the human brain could ever be replicated by a purely electrical device. One of the goals of AI research is to try to create a general intelligence, which seems out of reach to researchers currently. They can make software that classifies pictures by who painted them, or recognizes animals and such, but something that does true and generalized “thinking” is elusive.

When I connect this idea to the human brain, it doesn’t function purely in an electrical fashion. There is also chemical reactions that trigger it. What if this architecture of how the brain functions is critical to the operation of general intelligence. Some AI researchers seem to think if they can create generalized intelligence the era of scarcity would end. We would have robots that serve us that don’t need sleep, but if the idea of chemical reactions are needed in the architecture to create generalized intelligence this would defeat the purpose. The chemical reactions that take place in the brain are why we need rest, and what needs time to reverse after too much usage under certain tasks. If this was required for generalized artificial intelligence then those would have all the same issues as humans in this regard. Mental illness possibilities. Fatigue possibilities. Needing breaks. Needing vacations and more.

Both of these are perhaps the most complex research undertakings happening in the world. There is no clue on what the needed architecture for a generalized intelligence is, nor do we understand exactly how the human brain works. However, I do find it a good probability to bet on that if we did understand how the brain works, we would likely understand that a computer made of purely electrical signals can’t replicate it entirely. You would have to make some sort of tradeoff between fatigable and smarter, or less fatigable and less intelligent.

Now that that is out of the way. Here’s another point, when you are stuck, much as both sides of this research is, it may be because you don’t see the tradeoffs that need to be made. Your career may be stalled because you are deciding between money and calmness. Your move may be stalled because you can’t decide between rural, relaxed and lacking attractions, or urban, hectic and never-ending attractions. Your product may be stalled because you can figure out if you want to market it as cheap to the masses, or a premium option to the deep pockets.

Everything interesting, complex, and worth doing has unique trade offs. It’s better if we see them. If we don’t it’s time to look deeper.