Validating Non-trivial Semantic Properties of Autonomous Robots

0564526 - ÚI 2023 RIV CH eng C - Konferenční příspěvek (zahraniční konf.)
Wiedermann, Jiří - van Leeuwen, J.
Validating Non-trivial Semantic Properties of Autonomous Robots.
Philosophy and Theory of Artificial Intelligence 2021 Conference Proceedings. Cham: Springer, 2022 - (Magnani, L.), s. 91-104. Studies in Applied Philosophy, Epistemology and Rational Ethic, 63. ISBN 978-3-031-09152-0.
[PTAI 2021: Conference on Philosophy and Theory of Artificial Intelligence /4./. Gothenburg (SE), 27.09.2021-28.09.2021]
Grant ostatní: AV ČR(CZ) StrategieAV21/1
Program: StrategieAV
Institucionální podpora: RVO:67985807
Klíčová slova: Autonomous robots * Cyber-physical systems * Ethics of AI Semantic properties * Rice’s theorem * Robot modelling * Turing machines * Verification
Obor OECD: Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

A semantic property of autonomous robots is called non-trivial if some robots satisfy it and some do not, like adherence to rules of ethics or compliance with legal regulations. In order to study the validation problem for these properties, we model robots as cyber-physical systems with programmable control. Their behaviour is modelled by the infinite streams of interactions that they generate. We show that, under mild conditions, there can be no algorithmic method for deciding from a robot’s program whether it satisfies a given non-trivial semantic property or not. The result provides a compelling analogue to Rice’s theorem from classical computability theory, now for autonomous robots. We also show that no interactive verifiers of any kind whatsoever can exist for the problem. The results are fundamental to understanding the difficulty of validations in artificial intelligence.
Trvalý link: https://hdl.handle.net/11104/0336186