0000000000004944
AUTHOR
Riccardo Manzotti
AGI and Machine Consciousness
This review discusses some of main issues to be addressed to design a conscious AGI agent: the agent’s sense of the body, the interaction with the environment, the agent’s sense of time, the free will of the agent, the capability for the agent to have some form of experience, and finally the relationship between consciousness and creativity.
Conscious Machines: A Possibility? If So, How?
The scope of the paper is to encourage scientists and engineering to avoid to do what Einstein pointed out as being the hallmark of folly. Machine consciousness scholars must be brave enough to step out of the beaten path. There must be some big recurrent conceptual mistakes that prevent science and technology from addressing machine consciousness.
Physical integration: A causal account for consciousness
The issue of integration in neural networks is intimately connected with that of consciousness. In this paper, integration as an effective level of physical organization is contrasted with a methodological integrative approach. Understanding how consciousness arises out of neural processes requires a model of integration in just causal physical terms. Based on a set of feasible criteria (physical grounding, causal efficacy, no circularity and scaling), a causal account of physical integration for consciousness centered on joint causation is outlined.
Machine consciousness: A manifesto for robotics
Machine consciousness is not only a technological challenge, but a new way to approach scientific and theoretical issues which have not yet received a satisfactory solution from AI and robotics. We outline the foundations and the objectives of machine consciousness from the standpoint of building a conscious robot. © 2009 World Scientific Publishing Company.
Artificial Consciousness
“Artificial” or “machine” consciousness is the attempt to model and implement aspects of human cognition that are identified with the elusive and con- troversial phenomenon of consciousness. The chapter reviews the main trends and goals of artificial consciousness research, as environmental coupling, autonomy and resilience, phenomenal experience, semantics or intentionality of the first and sec- ond type, information integration, attention. The chapter also proposes a design for a general “consciousness oriented” architecture that addresses many of the discussed research goals. Comparisons with competing approaches are then presented.
Good Old-Fashioned Artificial Consciousness and the Intermediate Level Fallacy
Recently, there has been considerable interest and effort to the possibility to design and implement conscious robots, i.e., the chance that a robot may have subjective experiences. However, typical approaches as the global workspace, information integration, enaction, cognitive mechanisms, embodiment, i.e., the Good Old-Fashioned Artificial Consciousness, henceforth, GOFAC, share the same conceptual framework. In this paper, we discuss GOFAC's basic tenets and their implication for AI and Robotics. In particular, we point out the intermediate level fallacy as the central issue affecting GOFAC. Finally, we outline a possible alternative conceptual framework towards robot consciousness.