Paths to non-deterministic autonomy: practical and qualitative considerations towards a Hawking-Musk-esque nightmare

Current computational approaches, such as Artificial Intelligence, artificial neural networks, expert systems, fuzzy logic, fuzzy-cognitive maps, other rule-based approaches, etc., fundamentally do not lend themselves to building non-deterministic autonomous reasoning systems. Especially for AI, high hopes were raised more than 50 years ago, but AI has largely failed to deliver on its promises and still does. As such, the paper discusses different ingredients and approaches towards completely non-deterministic autonomous systems that are based on and exhibit critical capabilities, such as, but not limited to, self-organization, self-configuration, and self-adaptation. As such, any two initially identical autonomous systems will exhibit diverging and ultimately completely unpredictable developmental trajectories over time, once exposed to the same or similar environment, and even more so, once exposed to different environments.Edward & Maria Keonjian Endowment at the University of ArizonaThis item from the UA Faculty Publications collection is made available by the University of Arizona with support from the University of Arizona Libraries. If you have questions, please contact us at repository@u.library.arizona.edu