Comment by crackalamoo

> It’s not like language keeps inventing new logic at a rate we can’t keep with.

Maybe it does. Maybe, to a smart enough model, given its training on human knowledge so far, the next logical thing after "Sure, here's a technically and economically feasible cure for disease X" is in fact such a cure, or at least useful steps towards it.

I'm exaggerating, but I think the idea may hold true. It might be too early to tell one way or another definitively.