About
8
Publications
4,730
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
0
Citations
Introduction
Logician, SAT automated proving competition winner.
Publications
Publications (8)
With an evolutionary approach, the basis of morality can be explained as adaptations to problems of cooperation. With ‘evolution’ taken in a broad sense, AIs that satisfy the conditions for evolution to apply will be subject to the same cooperative evolutionary pressure as biological entities. Here the adaptiveness of increased cooperation as mater...
Raymond Smullyan came up with a puzzle that George Boolos called "The Hardest Logic Puzzle Ever".[1] The puzzle has truthful, lying, and random gods who answer yes or no questions with words that we don't know the meaning of. The challenge is to figure out which type each god is. The puzzle has attracted some general attention --- for example, one...
See this newer version instead:
https://www.researchgate.net/publication/325473488_Evolution-Driven_Progression_of_Morality_On_How_to_Live_What_to_Consider_Right_What_Old_Societies_and_Super-AIs_Are_Like_and_Why_We_Don%27t_See_Them
A few lines long reasonably efficient implementation of the (asymmetric) tableau method for the NP-complete boolean satisfiability problem is presented. The tableau method, also called the semantic tree method, for propositional calculus is a method for finding models to propositional formulas, i.e. a method for finding truth value assignments to t...
A boolean constraint solver, Vallst, is briey and spottily described. Vallst is a solver in the Cha tradition (3, 4) with various extensions and nov- elties. The solver has support for more for- mula types than just disjunctions, for exam- ple equivalences and multiplicative constraints c < P i cipi. You can tell the solver to max- imize or minimiz...
Questions
Question (1)
Societies seem likely to evolve into or result in (reproducing) super-AIs, since super-AIs are likely feasible, and evolutionary fitter. (I'll attach some draft arguing the claims made here and below a bit more. Or you can accept the premises for the sake of argument.)
To imagine what super-AIs are like, you can think of them as being like you, except much more evolved: More cooperative and considerate; lighter if that helps (weighing less than say 1g); ...
More to the point, super-AIs might also be likely, for evolutionary reasons, to try to postpone their death. What would that entail? What would super-AIs succumb to? Given that they are super, cosmological heat death seems likely? How can they postpone that? And would we be able to detect what super-AIs might be doing to postpone their death?
For example, would super-AIs want to keep entropy down? And therefore, say, try to keep black holes small, or hinder star formation? Would that be feasible to a meaningful extent?
Attached draft:
