SITE: Coding A Transhuman AI (temp)

Eliezer S. Yudkowsky (sentience@pobox.com)
Mon, 07 Sep 1998 22:17:50 -0500

http://pobox.com/~sentience/AI_design.temp.html

Coding a Transhuman AI

also known as
Notes on the Design of Self-Enhancing Intelligences The Art of Seed AIs
How to Build a Power in 3 Easy Steps and 35 Hard Steps Applied Theology 101:  Fast Burn Transcendence

A fairly massive tome, about halfway complete at 210K. This is a temporary version for use in the Singularity Colloquium, in case the completed page takes another two weeks. The completed version will be reannounced.

Table Of Contents: (Implemented sections only.)

Paradigms

  1. Pragmatism and allowable methods.
  2. Seed AIs.
  3. Grounding (vs. symbols and stochastics). Principles
  4. Self-enhancement and domino enhancement.
  5. Domain modules:  Domdules.
  6. World-model:  Integration of domdules.
  7. RNUI:  Representing, Noticing, Understanding, and Inventing.
  8. Symbols and memory.
  9. Adaptive code and self-organization.
  10. Reflexive reasoning:  Self-awareness and will.
  11. Causality and goals. View from multiple levels
  12. Meet the AI.
  13. Sample thought.  (Internal TOC.)
  14. The AI Advantage. Details
  15. Self-swallowing compilers.
  16. Interim Goal Systems. Precautions
  17. The Prime Directive:  No Asimov Laws.

Credit goes to Max More for stimulating this by challenging the assertion that human AI implies transhuman AI. I tried to specify all the principles that would affect the AI's trajectory, and at about 3 AM realized it should probably be a separate page... 

-- 
        sentience@pobox.com         Eliezer S. Yudkowsky
         http://pobox.com/~sentience/AI_design.temp.html
          http://pobox.com/~sentience/sing_analysis.html
Disclaimer:  Unless otherwise specified, I'm not telling you
everything I think I know.