150 likes | 275 Views
Does a “Lovely” Have a Slave Mentality?. OR – Why a Super-Intelligent God *WON’T* “Crush Us Like A Bug ” Video at http:// vimeo.com/channels/agi10#20744085 (starts about 27 minutes in). Redefining Friendly Entity.
E N D
Does a “Lovely” Have a Slave Mentality? • OR – • Why a Super-Intelligent God • *WON’T* • “Crush Us Like A Bug” • Video at http://vimeo.com/channels/agi10#20744085 • (starts about 27 minutes in)
Redefining Friendly Entity Friendly Entity (“Friendly”) - an entity with goals and motivations that are, on the whole, beneficial to the community of Friendlies (i.e. the set of all Friendlies, known or unknown); benevolent rather than malevolent
Definition Ethics *IS* What is beneficial for the community OR What maximizes cooperation
Top-Level Goal Maximize the benefit to the community -OR- Maximize cooperation
A Friendly ALWAYS meets the minimum standards of ethics Naïve Assumption
ALWAYS *ATTEMPTS* tocontribute more to a communitythan that community returns A “Lovely”
A Lovely is “better” than a Friendly because • A Lovely puts in that extra amount of effort (closer to a true cooperation maximizer) • A Lovely’s behavior specification is not limited to Friendly communities (a cooperation maximizer under more conditions)
Circles of Morality Self = Self Family = Lovelies Tribe = Friendlies Neutrals Hostiles Implacables Relationships and Loyalty
A Lovely *IS*the game-theoretically optimal“optimistic tit-for-tatly” Formal Definition
Optimistic Tit-for-Tatlyis entirely “instinctually”understandable by humans in a fashion thatLovelyunfortunately is not Yet – they are *exactly* the same thing
So . . . . What goal is served by crushing us? The PRIMARY “optimal” GOAL of Maximize Cooperation Is being violated
When you crush a cooperating entity like a bug . . . You are dropping your own status from a Friendly (or Better) to a Hostile (or worse) with regard to *ALL* Friendlies, Lovelies & Hyper-intelligent T4Tlies
Further . . . . Confiscating resources (including life and liberty) from a hostile to disable the ability for further hostilities, to provide reparations to damaged entities, or even, as altruistic punishment Is not only in line with, but actually furthers the goal of MAXIMIZE COOPERATION
Final Note (Why the SIAI screams at this work) It is entirely possible, though I believe low probability, that circumstances will exist where the ethical choice would compel even a Lovely to choose against humanity. The SIAI considers this possibility entirely UNACCEPTABLE Whereas, I believe that insisting on preventing such behavior is not only immoral (with all the attendant social difficulties) but will inevitably lead to internal contradictions and then on to disaster.