3 ms·
in 2018 we had the first gtp that would babble and repeat itself but would string together words that were oddly coherent. people dismissed any talk of these mo
by AI_beffr 2y ago
in 2018 we had the first gtp that would babble and repeat itself but would string together words that were oddly coherent. people dismissed any talk of these models having larger potential. and here we are today with the state of AI being what it is and people are still, in essence, denying that AI could become more capable or intelligent than it is right at this moment. after so many years of this zombie argument having its head chopped off and then regrowing, i can only think that it is peoples deep seated humanity that prevents them from seeing the obvious. it would be such a deeply disgusting and alarming development if AI were to spring to life that most people, being good people, are literally incapable of believing that its possible. its their own mind, their human sensibilities, protecting them. thats ok. but it would help keep humanity safe if more people had the ability to realize that there is nothing stopping AI from crossing that threshold and every heuristic is pointing to the fact that we are on the cusp of that.