Re: [sl4] to-do list for strong, nice AI

From: Pavitra (
Date: Tue Oct 13 2009 - 00:29:04 MDT

Pavitra wrote:
> A[ ] Develop a mathematically formal definition of Friendliness.
> A->B[ ] Develop an automated test for Friendliness with a 0% false
> positive rate and a reasonably low false negative rate.
> C[ ] Develop a mathematically formal definition of intelligence.
> C->D[ ] Develop an automated comparison test that returns the more
> intelligent of two given systems.
> B,D->E[ ] Develop prototype systems and apply these tests to them
> iteratively until the Singularity occurs.

Comments, per request:

Wrote out dependencies more specifically; this should make clearer what
needs to be worked on and what can't be worked on yet.

Rewrote definition and test desiderata more rigorously and calling for
more rigor; this should make less ambiguous whether or not a given step
is completed yet.

Consolidated last two points into a single step (E); the alternatives
were circular dependencies (this was the existing situation), or an
indefinitely large number of steps.

Messed with whitespace a bit for readability.

Todo: Step E seems non-obvious to me. Why is iterated prototyping
necessarily the next step after B,D are accomplished?

This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:01:04 MDT