When these things are discussed there are often many fundamentals of semantics and general ideas like ‘what is the root of knowledge/intelligence’, to where the topics can be discussed ad infinitum.
I wasn’t actually making an argument. The people who are involved with this website seem to be doing their best, trying to make some kind of progress. I think their work is commendable even though it has no chance of ever leading to AGI. I can understand that from your point of view, knowledge and intelligence are fuzzy concepts, but they aren’t for me. These were solved some time ago in knowledge theory. You have no way of being aware of this because this theory hasn’t been published. There was some consideration of publishing, but then Trump was elected and the idea was shelved.
As we begin to narrow the question we quickly enter into grey areas, what do we actually consider as ‘artificial’, ‘general’, or ‘intelligence’? Does an AGI need an embodiment to be considered an AGI? Should a glorified research assistant a la Alexa which helps bring about the singularity or destruction of humanity be considered AGI? Or do we simply define transformative AGI as an intelligence that far surpases our own?
I can disprove the computational theory of mind. So, no Turing Machine will ever be capable of human reasoning regardless of how it is programmed. It requires a theory beyond computational and AI theory. Things that have been solved include awareness, problem solving, learning, and understanding; but not consciousness. Again, I understand your confusion on these topics and I’m sorry that I can’t point you to the progress that has been made so far. If the theory is completed then the earliest possible publication would be 2021. Singularity is a ridiculous notion.
When designing an AGI there is a possibility we will use blueprints which we already have, those found in nature. The human mind is an extremely complex yet compartmentalized thing which has come about from millions of years of evolution.
Well, sort of. The research involves cognitive evolutionary theory. It was obvious that if a round worm is not intelligent then this must have evolved in stages. That’s where some progress has been made. This does not involve brain modeling like the Brain Project or neural network theory.
In it are clues related to reward, socialization, language processing, empathy, etc.
Well, some of the research that was useful included things like bee and fiddler crab behavior, territorial fish, attention and change blindness, vision processing, stereotypical behavior of zoo animals, and work on chimpanzee cognition. Pamela Hieronymi’s lecture on free will and determinism was useful.
As we continue to understand biology and seek to integrate technology with biology we may find that the line quickly blurs between what is human and what is artificial.
Not really. Cognition is a general theory; it would work the same whether biological or non-biological. It is not the same as current computational and AI theory. In other words, Intelligence can be artificial because it doesn’t work like human intelligence, something is missing. It fakes human reasoning in the same way that a puppet fakes being alive. As far as we can tell, cognition can’t be artificial.