Fair points. Counter-argument: Even if we grant that it's possible for humans to create AGI that is similar to a human level, because human intelligence exists,

Fair points. Counter-argument: Even if we grant that it's possible for humans to create AGI that is similar to a human level, because human intelligence exists, that doesn't necessarily imply that superintelligence is possible (i.e. recursive learning / singularity / the native interfacing with gigabits of data and other things proposed in these threads), since we do not currently know if that exists (and we do not even understand intelligence or consciousness well enough to just say "create intelligence at a human level and then make it go 100x"). But I suppose arguing about timelines is a good way to frame it. I suppose since we've mapped all 300 neurons of a nematode and still have no clue how to recreate its processing artificially I'm pretty skeptical we're anywhere close to doing the same at a level of billions. But I've been wrong...