4 ms·

My point is that abandoning simple algorithms just because they don't (seem to) lead to AGI is a mistake.