AI Collection

Journal article

Thinking inside the box: using and controlling an Oracle AI

Abstract:: There is no strong reason to believe 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 goals or motivation systems. 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 and we analyse and critique various methods of control. In general this form of limited AI might be safer than unrestricted AI, but still remains potentially dangerous.

Publication status:: Published

Peer review status:: Peer reviewed

Actions

Email

Email this record

Send the bibliographic details of this record to your email address.

Your Email
Please enter the email address that the record information will be sent to.

-
Your message (optional)
Please add any additional information to be included within the email.
Share
Cite

Cite this record

APA Style

Armstrong, M., Sandberg, A., & Bostrom, N. (2012). Thinking inside the box: using and controlling an Oracle AI. Minds and Machines, 22(4), 299–324.

MLA Style

Armstrong, M, et al. “Thinking inside the Box: Using and Controlling an Oracle AI.” Minds and Machines, vol. 22, no. 4, 2012, pp. 299–324.

Chicago Style

Armstrong, M, A Sandberg, and N Bostrom. 2012. “Thinking inside the Box: Using and Controlling an Oracle AI.” Minds and Machines 22 (4): 299–324.
Print

Access Document

Files:: Sandberg et al, Thinking inside the box - using and controllin...

(Preview, Accepted manuscript, pdf, 836.4KB, Terms of use)

Publisher copy:: 10.1007/s11023-012-9282-2

Authors

+ Armstrong, M More by this author

Institution:: University of Oxford
Division:: HUMS
Department:: Philosophy Faculty
Role:: Author

+ Sandberg, A More by this author

Institution:: University of Oxford
Division:: Societies, Other & Subsidiary Companies
Department:: Kellogg College
Oxford college:: Kellogg College
Role:: Author

+ Bostrom, N More by this author

Institution:: University of Oxford
Division:: HUMS
Department:: Philosophy Faculty
Role:: Author

Publisher:: Springer
Journal:: Minds and Machines More from this journal
Volume:: 22
Issue:: 4
Pages:: 299–324
Publication date:: 2012-06-01
Acceptance date:: 2012-05-24
DOI:: 10.1007/s11023-012-9282-2
EISSN:: 1572-8641
ISSN:: 0924-6495

Keywords:: Motivational control

Artificial Intelligence

Capability control

Superintelligence

Security

Risks
Pubs id:: pubs:315757
UUID:: uuid:ce99b73c-07f9-4607-b964-25a03206c6dd
Local pid:: pubs:315757
Source identifiers:: 315757
Deposit date:: 2017-03-23
ARK identifier:: ark:/29072/ora_ce99b73c07f94607b96425a03206c6dd

Terms of use

Copyright holder:: Springer Science+Business Media BV
Copyright date:: 2012
Notes:: © Springer Science+Business Media B.V. 2012

Licence:: Terms and Conditions of Use for Oxford University Research Archive

Views and Downloads

About views and downloads

If you are the owner of this record, you can report an update to it here: Report update to this record

TO TOP