Levels of Self-Improvement in AI and their Implications for AI Safety

Abstract

Abstract: This article presents a model of self-improving AI in which improvement could happen on several levels: hardware, learning, code and goals system, each of which has several sublevels. We demonstrate that despite diminishing returns at each level and some intrinsic difficulties of recursive self-improvement—like the intelligence-measuring problem, testing problem, parent-child problem and halting risks—even non-recursive self-improvement could produce a mild form of superintelligence by combining small optimizations on different levels and the power of learning. Based on this, we analyze how self-improvement could happen on different stages of the development of AI, including the stages at which AI is boxed or hiding in the internet.

Links

PhilArchive

External links

  • This entry has no external links. Add one.
Setup an account with your affiliations in order to access resources via your University's proxy server

Through your library

  • Only published works are available at libraries.

Similar books and articles

Risks of artificial general intelligence.Vincent C. Müller (ed.) - 2014 - Taylor & Francis (JETAI).
Risks of artificial intelligence.Vincent C. Müller (ed.) - 2016 - CRC Press - Chapman & Hall.
Climate change, justice and the right to development.Lars Löfquist - 2011 - Journal of Global Ethics 7 (3):251-260.
Advantages of artificial intelligences, uploads, and digital minds.Kaj Sotala - 2012 - International Journal of Machine Consciousness 4 (01):275-291.
On Human Improvement.Asher Seidel - 1999 - International Journal of Applied Philosophy 13 (2):209-222.
School Improvement: Reality and Illusion.Robert Coe - 2009 - British Journal of Educational Studies 57 (4):363-379.

Analytics

Added to PP
2018-04-08

Downloads
583 (#30,705)

6 months
112 (#36,955)

Historical graph of downloads
How can I increase my downloads?

Author's Profile

Citations of this work

No citations found.

Add more citations

References found in this work

No references found.

Add more references