Autonomous AI agents can remain fully authorized and still become unsafe as behavior drifts, adversaries adapt, and decision patterns shift without any code change. We propose the \textbf{Informational Viability Principle}: governing an agent reduces to estimating a bound on unobserved risk $\hat{B}(x) = U(x) + SB(x) + RG(x)$ and allowing an action only when its capacity $S(x)$ exceeds $\hat{B}(x)$ by a safety margin. The \textbf{Agent Viability Framework}, grounded in Aubin's viability theory, establishes three properties -- monitoring (P1), anticipation (P2), and monotonic restriction (P3) -- as individually necessary and collectively sufficient for documented failure modes. \textbf{RiskGate} instantiates the framework with dedicated statistical estimators (KL divergence, segment-vs-rest $z$-tests, sequential pattern matching), a fail-secure monotonic pipeline, and a closed-loop Autopilot formalised as an instance of Aubin's regulation map with kill-switch-as-last-resort; a scalar Viability Index $VI(t) \in [-1,+1]$ with first-order $t^*$ prediction transforms governance from reactive to predictive. Contributions are the theoretical framework, the reference implementation, and analytical coverage against published agent-failure taxonomies; quantitative empirical evaluation is scoped as follow-up work.
Deontic Policies for Runtime Governance of Agentic AI Systems
CSEE Department UMBC Baltimore, MD, USA;Center for AI UMBC Baltimore, MD, USA;Information Systems Department UMBC Baltimore, MD, USA;CSAIL MIT Cambridge, MA, USA
Integrated Design and Governance of Agentic AI Systems through Adaptive Information Modulation
Department of Mechanical and Industrial Engineering(机械与工业工程系);Institute of Experiential AI(体验人工智能研究所);Network Science Institute(网络科学研究所);Northeastern University(东北大学);Boston, MA 02115, United States(波士顿,马萨诸塞州,02115,美国)
Think Before You Act -- A Neurocognitive Governance Model for Autonomous AI Agents
Old Dominion University(老奥德纳大学);AI Motion Labs(AI运动实验室);Department of Psychiatry and Neurobehavioral Sciences(精神病学与神经行为科学系);University of Virginia School of Medicine(弗吉尼亚大学医学院);Florida International University(佛罗里达国际大学);Accenture Technology Labs(埃森哲技术实验室);Center for Wireless Communications(无线通信中心);University of Oulu(奥卢大学);Blanchfield Army Community Hospital(布莱恩菲尔德陆军社区医院);McDonald Army Health Center(麦克唐纳陆军健康中心)
OpenPort Protocol: A Security Governance Specification for AI Agent Tool Access
Accentrust Georgia Institute of Technology(阿肯色州立大学);Accentrust University of Illinois Urbana-Champaign(阿肯色州立大学);Duke University(杜克大学);New York Institute of Technology-Vancouver(纽约理工学院-温哥华分校);Georgia Institute of Technology(佐治亚理工学院)
Hebrew University of Jerusalem(耶路撒冷希伯来大学);University of Toronto(多伦多大学);Schwartz Reisman Institute for Technology and Society(施瓦茨·里斯曼技术与社会研究所);Institute for Law & AI(法律与人工智能研究所)
Towards Governing Agent's Efficacy: Action-Conditional $β$-VAE for Deep Transparent Reinforcement Learning
Seoul National University(首尔国立大学);Samsung Electronics(三星电子)
我们在强化学习(RL)的设置中解决了深度神经网络的黑盒问题,其中神经代理以不可控的方式学习最大化奖励收益。当交互环境包括广阔的状态空间时,这种学习方法是有风险的,因为几乎不可能预见所有不想要的结果并预先用负奖励惩罚它们。与以前的作品中学习的神经功能的反向分析不同,我们提出的方法\nj{tackles the blackbox issue by encouraging}是一个RL策略网络,通过实现一个解纠缠的表示学习方法来学习可解释的潜在特征。为此,我们的方法允许RL代理通过将其影响与不可控的环境因素区分开来来理解自我效能,这与人类理解场景的方式非常相似。我们的实验结果表明,学习的潜在因素不仅是可解释的,而且还可以建模整个访问状态空间的分布与特定的动作条件。我们已经实验表明,拟议结构的这一特征可以导致对RL代理的期望行为进行事后治理。
EmbodiedGovBench: A Benchmark for Governance, Recovery, and Upgrade Safety in Embodied Agent Systems
School of Software, Harbin Institute of Technology(哈尔滨工业大学软件学院);School of Computer Science and Technology, Harbin Institute of Technology(哈尔滨工业大学计算机科学与技术学院);School of Mathematical and Computer Sciences, Heriot-Watt University, Malaysia Campus(赫瑞-瓦特大学马来西亚校区数学与计算机科学学院);School of Future Science and Engineering, Soochow University(苏州大学未来科学与工程学院)
Resilience Meets Autonomy: Governing Embodied AI in Critical Infrastructure
Department of Automation and Process Engineering (IAP)(自动化与过程工程系);UiT The Arctic University of Norway(北极大学);Department of Technology and Safety (ITS)(技术与安全系)