Fernando Rosas

• Fernando Rosas 13 Jun 2026 21:48 UTC

  1 point

  0

  in reply to: Winter Cross’s comment on: AI in a vat: Fun­da­men­tal limits of effi­cient world mod­el­ling for safe agent sandboxing

  You are absolutely right, thanks for spotting this! I just fixed it

• Fernando Rosas 17 Mar 2026 15:37 UTC

  1 point

  0

  on: On The In­de­pen­dence Axiom

  Great post! I have a small question about the sophisticated chooser. I see that it was clarified that it can potentially be money-pumped. Does that mean that it doesn’t satisfy dynamical consistency?

  What about consequentialism? Is that depending on if the local preferences depend or not on closed branches? (is that the definition of “locality”?)

• Fernando Rosas 3 Oct 2025 5:28 UTC

  1 point

  0

  in reply to: watermark’s comment on: AI in a vat: Fun­da­men­tal limits of effi­cient world mod­el­ling for safe agent sandboxing

  The point is that a coarse-graining can turn a quasi-probability into a probability. This is significant because quasi-probabilities are limited in that they cannot be sampled etc.

  Does that make sense?

AI in a vat: Fun­da­men­tal limits of effi­cient world mod­el­ling for safe agent sandboxing