Managed Autonomy at Runtime: Gear-Based Safety and Governance for Single- and Multi-Agent Cyber-Physical Systems

본문 미리보기

arXiv:2607.00334v1 Announce Type: new Abstract: Autonomous agents, whether LLM-driven software agents or robotic physical agents, face a common class of failure modes when operating without continuous human oversight: safety violations from unverified actions, behavioral instability from unconstrained loops, and continuity loss from unhandled error states. We develop \system{}, a discrete-time control system that combines five execution gears (\Gobs{}, \Gsug{}, \Gplan{}, \Gexec{}, \Gint{}) with

Managed Autonomy at Runtime: Gear-Based Safety and Governance for Single- and Multi-Agent Cyber-Physical Systems

본문 미리보기

관련 글

Constructive Alignment: Governing Preference Dynamics in Human-AI Interaction

Bounded Morality: Defining the Space of Moral Computation

The MMM Data Model -- A Normative Specification for Knowledge Interoperability in a Decentralisable Knowledge Commons

A Contextual-Bandit Oversight Game with Two-Sided Informational Asymmetry