Restricting the Model, Missing the System: Measurement and Accountability in Offensive AI Governance
AI Safety Department(人工智能安全部门);SimulaMet and OsloMet(SimulaMet和奥斯陆计量)
Michael Alexander Riegler, Finn Schwall, Annika Willoch Olstad, Birk Sebastian Frostelid Torpmann-Hagen, Sushant Gautam, Klas H. Pettersen, Inga Strümke
We show that the instruments used to measure AI offensive capability fail in two ways: (i) they overstate harm, and (ii) they credit the model with capability that belongs to the surrounding system. We argue that restricting access to a model is therefore necessary but not sufficient and that policy and procurement also need system-level, harm-grounded capability assessment. In June 2026, two frontier models were suspended under US export controls, reportedly prompted by a jailbreak that asked a model to read a codebase and fix its flaws. This finding measured an elicitation \emph{system} of model, prompt, and task. We support our argument with a study of an open-source framework in which lightweight large language model (LLM) agents coordinate through shared memory and evolutionary optimization, providing two pieces of evidence. First, jailbreak metrics overstate harm: over 225 swarm-generated attacks per target, LLM-as-judge scoring rated Claude Sonnet 4 compromised in 40% of attacks, yet manual verification found actionable harmful content in none, against a 45.8\% Effective Harm Rate for GPT-4o. Second, scaffolded evaluations misattribute capability: on a planted-vulnerability target, a full pipeline built around a 1.2B-parameter model recovers 9 of 9 weaknesses, while the same model without the hand-crafted components recovers 0 of 9 by crash verification and 2 of 9 by cited source line. Offensive capability is a property of model, scaffold, and evaluation protocol together. The duty to assess it should lie with whoever controls the system, shapes its behaviour, and can foresee what it will do. In our setting, that is the party who builds the harness around the model, a role that current regulation does not clearly cover.
A Framework for Responsible AI Systems: Building Societal Trust through Domain Definition, Trustworthy AI Design, Auditability, Accountability, and Governance
Deparment of Computer Science and Artificial Intelligence and Andalusian Research Institute in Data Science and Computational Intelligence (DaSCI), University of Granada(计算机科学与人工智能系和安达卢西亚数据科学与计算智能研究所(DaSCI),格拉纳达大学);TECNALIA (BRTA)(TECNALIA(BRTA));University of the Basque Country (UPV/EHU)(巴斯克大学(UPV/EHU));School of Engineering, Cornell University(工程学院,康奈尔大学);ADIA Lab(ADIA实验室);Computational Research Department, Lawrence Berkeley National Laboratory(计算研究部,劳伦斯伯克利国家实验室);Intelligent Systems for Robotics and Automation Laboratory, Macau University of Science and Technology(机器人与自动化智能系统实验室,澳门科学技术大学)
Data Obfuscation through Latent Space Projection (LSP) for Privacy-Preserving AI Governance: Case Studies in Medical Diagnosis and Finance Fraud Detection
Operational AI Deployment Assurance: Governance-State Orchestration Under Threshold-Sensitive Deployment Conditions -- A Governance Framework for High-Stakes AI Systems
Ducaltus | AI Assurance \& Governance Newcastle upon Tyne, United Kingdom School of Computing, Engineering \& Digital Technologies Teesside University Middlesbrough, United Kingdom
Human-AI Governance (HAIG): A Trust-Utility Approach
The Digital Statecraft Academy, United Kingdom(英国数字国家学府);Department of Computer Science, University College London (UCL), United Kingdom(英国伦敦大学学院计算机科学系)
A governance horizon for ethical-use constraints in open-weight AI models
School of Computer Science(计算机科学学院);Ministry of Education(教育部);Laboratory of High Confidence Software Technologies(高可信软件技术实验室);University of Science and Technology Beijing(北京科技大学);University of California, Davis(加州大学戴维斯分校)
通过审计Hugging Face Hub上的模型仓库,发现基于披露的治理在开放权重AI中具有浅层结构性限制,提出治理视野概念并比较不同政策设计的效果。
A federated architecture for sector-led AI governance: lessons from India
Department of Telecommunications, Ministry of Communications, Sanchar Bhawan, New Delhi, India(印度通信部电信局,新德里);Defence Institute of Advanced Technology, Ministry of Defence, Girinagar, Pune, India(印度国防部国防高级技术研究所,浦那)
Responsible AI Adoption in the Public Sector: A Data-Centric Taxonomy of AI Adoption Challenges
University of Tartu(塔尔图大学);University of Hradec Kralove(赫拉德茨-克拉洛韦大学);UAEM(墨西哥州自治大学);University of Twente(特文特大学);TU Delft(代尔夫特理工大学);University of Warsaw(华沙大学);Fortinet(飞塔公司)
Identity-Bound Governance Under Execution Uncertainty: An Accountability Proof Block for LLM Agent Persistent Halts, with Cryptographic Implementation and Cross-Model Calibration
Governance-Constrained Agentic AI: Blockchain-Enforced Human Oversight for Safety-Critical Wildfire Monitoring
AI Center, Faculty of Computer and Information Systems, Islamic University of Madinah(伊斯兰大学麦地那分校计算机与信息系统学院人工智能中心);Faculty of Computer Studies, Arab Open University-Bahrain(阿拉伯开放大学巴林分校计算机研究学院);Department of Computer Science, The Islamia University of Bahawalpur(巴哈瓦尔布尔伊斯兰大学计算机科学系);College of Computer Studies, University of Technology Bahrain(巴林科技大学计算机研究学院)
Never compromise with vulnerabilities: a comprehensive survey on AI governance
Institute of Artificial Intelligence (TeleAI), China Telecom(中国电信人工智能研究院(TeleAI));Beihang University(北京航空航天大学);Zhejiang University(浙江大学);Beijing University of Posts and Telecommunications(北京邮电大学);University of Chinese Academy of Sciences(中国科学院大学);University of Science and Technology of China(中国科学技术大学);University of Amsterdam(阿姆斯特丹大学);Beijing Institute of Technology(北京理工大学);University of Washington(华盛顿大学);Northwestern Polytechnical University(西北工业大学);Wuhan University(武汉大学);Shanghai Collaborative Innovation Center for AI Social Governance(上海人工智能社会治理协同创新中心);International Research Center for Complexity Sciences, Hangzhou International Innovation Institute, Beihang University(北京航空航天大学杭州创新研究院复杂科学国际研究中心);Singapore Management University(新加坡管理大学);CFAR, Technology and Research (A*STAR)(新加坡科技研究局计算、通信与风险中心);Shenzhen Campus of Sun Yat-sen University(中山大学深圳校区);National University of Singapore(新加坡国立大学);University of Barcelona(巴塞罗那大学)