"People are actively trying to make it happen" != "people trying can make it happen". People tried to make alchemy work, too - and they achieved some results along the way! But the foundation was wrong, and therefore all the work was not able to lead to the desired goal.
Is something GPT-like the right foundation for AGI? That is very far from proven.
I make no claims about any specific architectures, only that human intelligence isn't anything special and so far we've done a pretty good job at blowing past it in a bunch of domains.
Is something GPT-like the right foundation for AGI? That is very far from proven.