Exactly, I think the current crop of models is capable of solving a lot of non-first-world problems. Many of them don't need full AGI to solve, especially if we

Exactly, I think the current crop of models is capable of solving a lot of non-first-world problems. Many of them don't need full AGI to solve, especially if we start thinking outside Silicon Valley.