With smarter robots, come struggles and fear

This NYT article is worth a read:

Fears being fueled by Bostrom’s book “Superintelligence” as well as Elon Musk and Stephen Hawking’s remarks.

Example, so-so attempt at rebuttal (a bit tongue in check in part, but with some more serious undertones)

Attempts to dispel the threat are hard because:
species and civilizations come and go, for many reasons
every technology can be used for good and bad
this time may be truly different, after the threshold where technology re-builds itself, on exponential change curves

The arguments to address the fear, so far have been one or combination of:
the threat is still far off, and short term benefits outweigh long-term risks – we can manage this
we can probably design friendly AIs – we can manage this (Bostrom’s plea/resolution at end of his book)
we will co-evolve with the new tools – augmented intelligence (“Advanced Chess” example of teams and tools)
we will merge with the new tools (Kurzweil – creepy augmented intelligence)
something else is more likely to kill us off first, and we need AI to work on those complex urgent problems

Leave a Reply