I just can’t see a future where the frontier labs (or their investors) would allow a treaty or even a priesthood. It feels too late to put the cat back in the bag.
Wow, this deep dive into the alignment problem is truely insightful! Do you think we can ever fully model these deep learning networks to mitigate risk?
I just can’t see a future where the frontier labs (or their investors) would allow a treaty or even a priesthood. It feels too late to put the cat back in the bag.
"let's ban things so that every other entity that doesn't agree to the ban will progress fastest and have every advantage" - genius
Wow, this deep dive into the alignment problem is truely insightful! Do you think we can ever fully model these deep learning networks to mitigate risk?