Whether current or near-term AI systems could be conscious is a topic of
scientific interest and increasing public concern. This report argues for, and
exemplifies, a rigorous and empirically grounded approach to AI consciousness:
assessing existing AI systems in detail, in light of our best-supported
neuroscientific theories of consciousness. We survey several prominent
scientific theories of consciousness, including recurrent processing theory,
global workspace theory, higher-order theories, predictive processing, and
attention schema theory. From these theories we derive "indicator properties"
of consciousness, elucidated in computational terms that allow us to assess AI
systems for these properties. We use these indicator properties to assess
several recent AI systems, and we discuss how future systems might implement
them. Our analysis suggests that no current AI systems are conscious, but also
suggests that there are no obvious technical barriers to building AI systems
which satisfy these indicators.
Beschreibung
[2308.08708] Consciousness in Artificial Intelligence: Insights from the Science of Consciousness
%0 Generic
%1 butlin2023consciousness
%A Butlin, Patrick
%A Long, Robert
%A Elmoznino, Eric
%A Bengio, Yoshua
%A Birch, Jonathan
%A Constant, Axel
%A Deane, George
%A Fleming, Stephen M.
%A Frith, Chris
%A Ji, Xu
%A Kanai, Ryota
%A Klein, Colin
%A Lindsay, Grace
%A Michel, Matthias
%A Mudrik, Liad
%A Peters, Megan A. K.
%A Schwitzgebel, Eric
%A Simon, Jonathan
%A VanRullen, Rufin
%D 2023
%K "Artificial "Computers "Machine "Neurons Cognition" Intelligence" Learning" Society" and
%T Consciousness in Artificial Intelligence: Insights from the Science of
Consciousness
%U http://arxiv.org/abs/2308.08708
%X Whether current or near-term AI systems could be conscious is a topic of
scientific interest and increasing public concern. This report argues for, and
exemplifies, a rigorous and empirically grounded approach to AI consciousness:
assessing existing AI systems in detail, in light of our best-supported
neuroscientific theories of consciousness. We survey several prominent
scientific theories of consciousness, including recurrent processing theory,
global workspace theory, higher-order theories, predictive processing, and
attention schema theory. From these theories we derive "indicator properties"
of consciousness, elucidated in computational terms that allow us to assess AI
systems for these properties. We use these indicator properties to assess
several recent AI systems, and we discuss how future systems might implement
them. Our analysis suggests that no current AI systems are conscious, but also
suggests that there are no obvious technical barriers to building AI systems
which satisfy these indicators.
@misc{butlin2023consciousness,
abstract = {Whether current or near-term AI systems could be conscious is a topic of
scientific interest and increasing public concern. This report argues for, and
exemplifies, a rigorous and empirically grounded approach to AI consciousness:
assessing existing AI systems in detail, in light of our best-supported
neuroscientific theories of consciousness. We survey several prominent
scientific theories of consciousness, including recurrent processing theory,
global workspace theory, higher-order theories, predictive processing, and
attention schema theory. From these theories we derive "indicator properties"
of consciousness, elucidated in computational terms that allow us to assess AI
systems for these properties. We use these indicator properties to assess
several recent AI systems, and we discuss how future systems might implement
them. Our analysis suggests that no current AI systems are conscious, but also
suggests that there are no obvious technical barriers to building AI systems
which satisfy these indicators.},
added-at = {2023-09-22T17:19:08.000+0200},
author = {Butlin, Patrick and Long, Robert and Elmoznino, Eric and Bengio, Yoshua and Birch, Jonathan and Constant, Axel and Deane, George and Fleming, Stephen M. and Frith, Chris and Ji, Xu and Kanai, Ryota and Klein, Colin and Lindsay, Grace and Michel, Matthias and Mudrik, Liad and Peters, Megan A. K. and Schwitzgebel, Eric and Simon, Jonathan and VanRullen, Rufin},
biburl = {https://www.bibsonomy.org/bibtex/2b5fb720cc8ba5ba2359ea08858fcbd3a/gharbeia},
description = {[2308.08708] Consciousness in Artificial Intelligence: Insights from the Science of Consciousness},
interhash = {416fcf95f11e4cc5deac175029cb0730},
intrahash = {b5fb720cc8ba5ba2359ea08858fcbd3a},
keywords = {"Artificial "Computers "Machine "Neurons Cognition" Intelligence" Learning" Society" and},
note = {cite arxiv:2308.08708},
timestamp = {2023-09-22T17:19:08.000+0200},
title = {Consciousness in Artificial Intelligence: Insights from the Science of
Consciousness},
url = {http://arxiv.org/abs/2308.08708},
year = 2023
}