AGI and the Knight-Darwin Law: why idealized AGI reproduction requires collaboration
In International Conference on Artificial General Intelligence. Springer (forthcoming)
Abstract
Can an AGI create a more intelligent AGI? Under idealized assumptions, for a certain theoretical type of intelligence, our answer is: “Not without outside help”. This is a paper on the mathematical structure of AGI populations when parent AGIs create child AGIs. We argue that such populations satisfy a certain biological law. Motivated by observations of sexual reproduction in seemingly-asexual species, the Knight-Darwin Law states that it is impossible for one organism to asexually produce another, which asexually produces another, and so on forever: that any sequence of organisms (each one a child of the previous) must contain occasional multi-parent organisms, or must terminate. By proving that a certain measure (arguably an intelligence measure) decreases when an idealized parent AGI single-handedly creates a child AGI, we argue that a similar Law holds for AGIs.Author's Profile
My notes
Similar books and articles
Measuring the intelligence of an idealized mechanical knowing agent.Samuel Alexander - 2020 - Lecture Notes in Computer Science 12226.
Infinite graphs in systematic biology, with an application to the species problem.Samuel A. Alexander - 2013 - Acta Biotheoretica 61 (2):181--201.
The role of language in intelligence.Daniel C. Dennett - 1994 - In Jean Khalfa (ed.), What is Intelligence? The Darwin College Lectures. Cambridge: Cambridge University Press.
Darwin, Faith, and Critical Intelligence.John F. Haught - 2011 - Proceedings of the American Catholic Philosophical Association 85:59-68.
Darwin, Faith, and Critical Intelligence.John F. Haught - 2011 - Proceedings of the American Catholic Philosophical Association 85:59-68.
Intelligence, Artificial and Otherwise.Paul Dumouchel - 2019 - Forum Philosophicum: International Journal for Philosophy 24 (2):241-258.
Practice Effects in Intelligence Tests.Knight Dunlap & Agnes Snyder - 1920 - Journal of Experimental Psychology 3 (5):396.
Too many instincts: contrasting philosophical views on intelligence in humans and non-humans.Susan G. Sterrett - unknown
Machine intelligence and the long-term future of the human species.Tom Stonier - 1988 - AI and Society 2 (2):133-139.
The singularity: A philosophical analysis.David J. Chalmers - 2010 - Journal of Consciousness Studies 17 (9-10):9 - 10.
Charles Darwin’s Theory of Moral Sentiments: What Darwin’s Ethics Really Owes to Adam Smith.Greg Priest - 2017 - Journal of the History of Ideas 78 (4):571-593.
Analytics
Added to PP
2020-05-09
Downloads
352 (#33,262)
6 months
42 (#32,246)
2020-05-09
Downloads
352 (#33,262)
6 months
42 (#32,246)
Historical graph of downloads
Author's Profile
Citations of this work
Measuring Intelligence and Growth Rate: Variations on Hibbard's Intelligence Measure.Samuel Alexander & Bill Hibbard - 2021 - Journal of Artificial General Intelligence 12 (1):1-25.
Short-circuiting the definition of mathematical knowledge for an Artificial General Intelligence.Samuel Alexander - forthcoming - Lecture Notes in Computer Science.
Can reinforcement learning learn itself? A reply to 'Reward is enough'.Samuel Allen Alexander - forthcoming - CIFMA 2021.