提出分层可控的AI执行框架，解决企业级AI缺乏监管与容错的问题。
Beyond Autonomy: A Dynamic Tiered AgentRunner Framework for Governable and Resilient Enterprise AI Execution
- 按任务风险动态分配资源和审核强度，平衡安全与效率
- 四类独立代理分工协作，物理隔离确保权责分离
- 故障自动触发恢复机制，系统设计自带韧性
当前大型语言模型代理框架过度追求自主性，缺乏企业部署所需的可管控机制。高风险写操作未经独立审查，复杂任务缺少结果验证，计算资源分配未区分风险等级。本文提出动态分层AgentRunner框架，源自生产级多租户SaaS平台。该框架引入三项核心机制：(1) 风险自适应分层，根据任务风险动态调整计算资源与审查强度，实现安全与效率的帕累托最优；(2) 权力分立架构，提案、审查、执行与验证由独立代理完成，具备物理隔离边界；(3) 设计即韧性，通过验证器-恢复闭环将故障视为系统第一状态。我们形式化了层级选择策略，并在真实企业工作流中验证其有效性。
原文摘要 · Abstract (English)
Current large language model agent frameworks prioritize autonomy but lack the governability mechanisms required for enterprise deployment. High-risk write operations proceed without independent review, complex tasks lack acceptance verification, and computational resources are allocated uniformly regardless of risk level. We propose the Dynamic Tiered AgentRunner, a controlled execution protocol distilled from a production-grade multi-tenant SaaS platform. The framework introduces three core mechanisms: (1) Risk-Adaptive Tiering that dynamically allocates computational resources and review intensity based on task risk profiles, achieving Pareto-optimal trade-offs between safety and efficiency; (2) Separation of Powers architecture where proposal, review, execution, and verification are performed by independent agents with physically isolated boundaries; and (3) Resilience-by-Design through a Verifier-Recovery closed loop that treats failure as a first-class system state. We formalize the tier selectio
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。