Deep learning is very adhoc and experimental right now. Anyone reading through the literature would see that. That's why I'm pessimistic about AGI, but very opt

Deep learning is very adhoc and experimental right now. Anyone reading through the literature would see that. That's why I'm pessimistic about AGI, but very optimistic about short-mid-term industry applications.