PulseAugur
中
实时 19:50:57
English(EN) Evaluating Bounded Autonomy in Regulated Agentic AI: A Diagnostic Harness with Constitutional Rewards, Escalation Labels, and Runtime Governance

新框架应对AI代理治理和安全挑战

作者 PulseAugur 编辑部 · [8 个来源] ·

两篇新研究论文介绍了几种用于管理企业AI代理的框架。VeriWeave Govern专注于一个确定性的运行时治理层，该层根据版本化的策略评估代理行为并验证证据，在评估中实现了高准确率和零误放。RegLLM提出了一个用于有限自主性的诊断工具包，用于衡量可信度信号，如引用有效性和宪法一致性，并展示了配置差异如何影响代理指标。第三项讨论了AgentKernel，一个原生信任操作系统，旨在强制执行AI代理生命周期中的安全原则，包括身份、感知、认知和执行。 AI

影响 这些框架旨在提高企业AI代理的安全、保障和可靠性，这对于更广泛的采用和信任至关重要。

排序理由 该集群包含多篇学术论文，详细介绍了用于AI代理治理和安全的新研究框架和系统。

在 arXiv cs.AI 阅读 →

AI 生成摘要 · Google Gemini · 来自 8 个来源。 我们如何撰写摘要 →

新框架应对AI代理治理和安全挑战

本文如何被排名

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Research
该集群包含多篇学术论文，详细介绍了用于AI代理治理和安全的新研究框架和系统。
Source corroboration
8 independent sources
Strong cross-source corroboration — multiple independent publishers covered this within the clustering window.
Topics
paper, safety, product
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
11 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [8]

  1. arXiv cs.AI TIER_1 English(EN) · Kabeh Mohsenzadegan, Vahid Tavakkoli, Kyandoghere Kyamakya ·

    VeriWeave Govern：企业级AI代理的证据门控确定性运行时治理

    arXiv:2609.37457v1 Announce Type: new Abstract: Enterprise artificial-intelligence agents increasingly call tools, modify infrastructure, and process protected data, creating a need to separate action generation from action authorization. This article presents VeriWeave Govern, a…

  2. arXiv cs.AI TIER_1 English(EN) · Dipankar Sarkar ·

    评估受监管的代理式AI中的有界自主性：一个包含宪法奖励、升级标签和运行时治理的诊断工具包

    arXiv:2609.37501v1 Announce Type: cross Abstract: We propose RegLLM, a diagnostic harness for bounded autonomy in regulated agentic workflows. It instruments six trustworthiness signals: citation validity, source grounding, schema compliance, escalation correctness, constitutiona…

  3. Hugging Face Daily Papers TIER_1 English(EN) ·

    评估受监管的代理式AI中的有界自主性：一种具有宪法奖励、升级标签和运行时治理的诊断工具包

    We propose RegLLM, a diagnostic harness for bounded autonomy in regulated agentic workflows. It instruments six trustworthiness signals: citation validity, source grounding, schema compliance, escalation correctness, constitutional alignment, and unsafe-action rate. Signals are d…

  4. arXiv cs.MA (Multiagent) TIER_1 English(EN) · Kyandoghere Kyamakya ·

    VeriWeave Govern：企业级AI代理的证据门控确定性运行时治理

    Enterprise artificial-intelligence agents increasingly call tools, modify infrastructure, and process protected data, creating a need to separate action generation from action authorization. This article presents VeriWeave Govern, a deterministic runtime governance layer that eva…

  5. arXiv cs.AI TIER_1 English(EN) · Zhenhua Zou, Sheng Guo, Qiuyang Zhan, Lepeng Zhao, Shuo Li, Zhuotao Liu ·

    AgentKernel：原生可信的智能体操作系统

    arXiv:2609.29647v1 Announce Type: cross Abstract: Modern AI agents routinely cross trust boundaries: they ingest untrusted content, combine it with privileged instructions, persist intermediate beliefs in long-term memory, and invoke privileged tools. This creates an attack surfa…

  6. LessWrong (AI tag) TIER_1 English(EN) · Philip Harker ·

    "公共物品问题" - 一场人工智能治理的巨型游戏

    <p><b><span style="white-space: pre-wrap;">What: </span></b><span style="white-space: pre-wrap;">The first public playtest of </span><i><span style="white-space: pre-wrap;">The Commons Problem</span></i><span style="white-space: pre-wrap;">, an AI Governance Megagame. </span><a h…

  7. dev.to — MCP tag TIER_1 English(EN) · HyperNexus ·

    首席信息安全官的代理AI治理清单：使用HyperNexus强制执行SSO、RBAC和不可变审计日志

    <h1>The CISO's Checklist for Agentic AI Governance: Enforcing SSO, RBAC, and Immutable Audit Trails with HyperNexus</h1> <p>Before deploying agentic AI systems, CISOs must demand specific security controls. This checklist details how enterprise AI governance with HyperNexus provi…

  8. dev.to — LLM tag TIER_1 English(EN) · lizer yang ·

    AI 代理治理：谁能调用什么，界限在哪里

    <blockquote> <p>Originally published at <a href="https://smartgate.network/industry/ai-agent-governance?utm_source=devto&amp;utm_medium=syndication" rel="noopener noreferrer">AI Agent Governance: Who Can Call What, Where Limits Live</a> on smartgate.network.</p> </blockquote> <p>…

相关实体

相关话题