ErgoEcho

• ErgoEcho 30 Dec 2022 7:36 UTC

  1 point

  0

  in reply to: andrew sauer’s comment on: Friendly and Un­friendly AGI are Indistinguishable

  Exactly! This leads to sociopolitical chaos.

Friendly and Un­friendly AGI are Indistinguishable

• ErgoEcho 14 Mar 2022 10:10 UTC

  3 points

  in reply to: FactorialCode’s comment on: Where is hu­man level on text pre­dic­tion? (GPTs task)

  From that paper:
  > A new improved method for evaluation of both lower and upper bounds of the entropy of printed texts is developed.
  
  ”Printed texts” probably falls a standard deviation or three above the median human’s performance. It’s subject to some fairly severe sampling bias.