Robotic agents with artificial intelligence may become empowered with moral reasoning and decision making capacities. Trumpeted by enthusiasts, dismissed by doubters, and feared by doomsayers, the implications of advancing machine intelligence should be thoroughly and honestly engaged.
Wallach outlines three key aspects of general artificial intelligence: complexity, thresholds and bioethics. Each hinges on the degree to which humans are similar or different to artificial systems. Complexity models equate microprocessing capacity to human cognition as quantified by the number of neural synapses. Moore's law makes these models compelling but current computational theories may not account for the complexity of brain mechanisms which are characterized by plasticity, massively parallel processing, and intricate feedback loops. Giving recent strides in visualization and cognitive science their due, we are still far from a true understanding of the mind and consciousness.
Do we want computers making moral decisions? If so, who defines morality? How can we make ethics computable at all? Wallach explores both top down pathways (such as Asimov's three laws of robotics) and bottom up pathways (such as evolutionary approaches) which may shape machine ethics.
Ultimately, he argues, we must keep in mind that we are biochemical platforms. Our intelligence emerged out of emotions and instincts. In contrast, computers start as logical platforms. In order for machines to to make decisions in harmony with humans, we may need to introduce emotional capacities, social skills, a sense of embodiment, and a theory of mind. Even so, right now we are best off recognizing the limits of what we can do.
Wendell Wallach is a lecturer and consultant at Yale University'sInterdisciplinary Center for Bioethics. Before coming to Yale, Wendellwas a founder and the President of two computer consulting companies,Farpoint Solutions and Omnia Consulting Inc. Among the clients servedby Mr. Wallach's companies were PepsiCo International, UnitedAircraft, and the State of Connecticut. At Yale University, Wendellchairs the working research group on Technology and Ethics, leads aseminar for bioethics interns, and functions as a senior coordinatorfor other working groups and projects. He has lectured worldwide,published many articles, and is presently writing two books.Cybersoul explores the ways in which cognitive science and theInformation Age are altering our understanding of humandecision-making and ethics. Machine Morality: From Aristotle toAsimov and Beyond, which Wendell is co-authoring and which will bepublished by MIT Press, explores the prospects for designing computersystems capable of making moral decisions. Wendell is recognized asone of the leaders in the new field of Machine Ethics, and designedthe first course anywhere on this subject, which he has taught twiceat Yale.
This free podcast is from our Singularity Summit series.