智能AI委派

摘要

人工智慧代理能夠處理日益複雜的任務。為實現更宏大的目標，AI代理需要具備將問題有效分解為可管理子組件的能力，並能安全地將這些子任務委派給其他AI代理或人類協作者。然而，現有的任務分解與委派方法依賴簡單啟發式規則，無法動態適應環境變化，也難以穩健處理意外故障。本文提出一種自適應的智能AI委派框架——該決策序列包含任務分配機制，同時整合權限轉移、職責劃分、問責體系、角色邊界明確定義、意圖清晰化以及建立多方信任的機制。所提出的框架適用於複雜委派網絡中的人類與AI委派方/受託方，旨在為新興智能代理網絡的協議開發提供理論基礎。

English

AI agents are able to tackle increasingly complex tasks. To achieve more ambitious goals, AI agents need to be able to meaningfully decompose problems into manageable sub-components, and safely delegate their completion across to other AI agents and humans alike. Yet, existing task decomposition and delegation methods rely on simple heuristics, and are not able to dynamically adapt to environmental changes and robustly handle unexpected failures. Here we propose an adaptive framework for intelligent AI delegation - a sequence of decisions involving task allocation, that also incorporates transfer of authority, responsibility, accountability, clear specifications regarding roles and boundaries, clarity of intent, and mechanisms for establishing trust between the two (or more) parties. The proposed framework is applicable to both human and AI delegators and delegatees in complex delegation networks, aiming to inform the development of protocols in the emerging agentic web.

智能AI委派

Intelligent AI Delegation

摘要

Support