Minds and Machines 22 (4):299-324 (2012)
Authors |
|
Abstract |
There is no strong reason to believe that human-level intelligence represents an upper limit of the capacity of artificial intelligence, should it be realized. This poses serious safety issues, since a superintelligent system would have great power to direct the future according to its possibly flawed motivation system. Solving this issue in general has proven to be considerably harder than expected. This paper looks at one particular approach, Oracle AI. An Oracle AI is an AI that does not act in the world except by answering questions. Even this narrow approach presents considerable challenges. In this paper, we analyse and critique various methods of controlling the AI. In general an Oracle AI might be safer than unrestricted AI, but still remains potentially dangerous.
|
Keywords | Artificial intelligence Superintelligence Security Risks Motivational control Capability control |
Categories | (categorize this paper) |
ISBN(s) | |
DOI | 10.1007/s11023-012-9282-2 |
Options |
![]() ![]() ![]() ![]() |
Download options
References found in this work BETA
Judgment Under Uncertainty: Heuristics and Biases.Daniel Kahneman, Paul Slovic & Amos Tversky (eds.) - 1982 - Cambridge University Press.
Judgment Under Uncertainty: Heuristics and Biases.Amos Tversky & Daniel Kahneman - 1974 - Science 185 (4157):1124-1131.
Theory of Games and Economic Behavior.John Von Neumann & Oskar Morgenstern - 1944 - Princeton, NJ, USA: Princeton University Press.
Judgement Under Uncertainty: Heuristics and Biases.Daniel Kahneman, Paul Slovic & Amos Tversky - 1985 - British Journal for the Philosophy of Science 36 (3):331-340.
View all 28 references / Add more references
Citations of this work BETA
Is It Time for Robot Rights? Moral Status in Artificial Entities.Vincent C. Müller - 2021 - Ethics and Information Technology 23 (3):579–587.
An AGI Modifying Its Utility Function in Violation of the Strong Orthogonality Thesis.James D. Miller, Roman Yampolskiy & Olle Häggström - 2020 - Philosophies 5 (40):40-0.
Racing to the Precipice: A Model of Artificial Intelligence Development.Stuart Armstrong, Nick Bostrom & Carl Shulman - 2016 - AI and Society 31 (2):201-206.
Human-Aligned Artificial Intelligence is a Multiobjective Problem.Peter Vamplew, Richard Dazeley, Cameron Foale, Sally Firmin & Jane Mummery - 2018 - Ethics and Information Technology 20 (1):27-40.
View all 14 citations / Add more citations
Similar books and articles
Releasing Philosophy, Thinking Art: A Phenomenological Study of Sylvia Plath's Poetry.Ellen Miller - 2007 - Davies Group, Publishers.
Trapped Inside the Box? Five Questions for Ben Fine.Michael A. Lebowitz - 2010 - Historical Materialism 18 (1):131-149.
Living Large: From SUVs to Double Ds---Why Going Bigger Isn't Going Better.Sarah Z. Wexler - 2010 - St. Martin's Press.
The Delphic Oracle H. W. Parke: A History of the Delphic Oracle. Pp. Viii+457; 8 Plates. Oxford: Blackwell, 1939. Cloth, 21s. [REVIEW]A. W. Gomme - 1940 - The Classical Review 54 (03):158-159.
A Naturalized Account of the Inside-Outside Dichotomy.Alvaro Moreno & Xabier Barandiaran - 2004 - Philosophica 73:11-26.
The Placement of Shadows : What's Inside William Kentridge's Black Box/Chambre Noire?Frances Guerin - 2011 - In John David Rhodes & Elena Gorfinkel (eds.), Taking Place: Location and the Moving Image. University of Minnesota Press.
From Bourdieu and Wolin, `Inside and Outside the Box': A Frame for the Special Issue.Lynda Stone & Michael Gunzenhauser - 2001 - Studies in Philosophy and Education 20 (3):181-190.
The Oracle at Didyma Joseph Fontenrose: Didyma. Apollo's Oracle, Cult and Companions. Pp. Xxi + 282; 3 Maps and 11 Illustrations. Berkeley, Los Angeles and London: University of California Press, 1988. $40. [REVIEW]Robert Parker - 1989 - The Classical Review 39 (02):270-271.
Why You Don’T Want to Get in the Box with Schrödinger's Cat.David Papineau - 2003 - Analysis 63 (1):51–58.
Analytics
Added to PP index
2012-06-06
Total views
257 ( #42,509 of 2,506,410 )
Recent downloads (6 months)
7 ( #102,990 of 2,506,410 )
2012-06-06
Total views
257 ( #42,509 of 2,506,410 )
Recent downloads (6 months)
7 ( #102,990 of 2,506,410 )
How can I increase my downloads?
Downloads