From Craft to Constitution: A Governance-First Paradigm for Principled Agent Engineering

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Xu, Qiang, Wen, Xiangyu, Xu, Changran, Li, Zeju, Zhong, Jianyuan
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908596591132672
author Xu, Qiang
Wen, Xiangyu
Xu, Changran
Li, Zeju
Zhong, Jianyuan
author_facet Xu, Qiang
Wen, Xiangyu
Xu, Changran
Li, Zeju
Zhong, Jianyuan
contents The advent of powerful Large Language Models (LLMs) has ushered in an ``Age of the Agent,'' enabling autonomous systems to tackle complex goals. However, the transition from prototype to production is hindered by a pervasive ``crisis of craft,'' resulting in agents that are brittle, unpredictable, and ultimately untrustworthy in mission-critical applications. This paper argues this crisis stems from a fundamental paradigm mismatch -- attempting to command inherently probabilistic processors with the deterministic mental models of traditional software engineering. To solve this crisis, we introduce a governance-first paradigm for principled agent engineering, embodied in a formal architecture we call ArbiterOS.
format Preprint
id arxiv_https___arxiv_org_abs_2510_13857
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle From Craft to Constitution: A Governance-First Paradigm for Principled Agent Engineering
Xu, Qiang
Wen, Xiangyu
Xu, Changran
Li, Zeju
Zhong, Jianyuan
Software Engineering
Artificial Intelligence
The advent of powerful Large Language Models (LLMs) has ushered in an ``Age of the Agent,'' enabling autonomous systems to tackle complex goals. However, the transition from prototype to production is hindered by a pervasive ``crisis of craft,'' resulting in agents that are brittle, unpredictable, and ultimately untrustworthy in mission-critical applications. This paper argues this crisis stems from a fundamental paradigm mismatch -- attempting to command inherently probabilistic processors with the deterministic mental models of traditional software engineering. To solve this crisis, we introduce a governance-first paradigm for principled agent engineering, embodied in a formal architecture we call ArbiterOS.
title From Craft to Constitution: A Governance-First Paradigm for Principled Agent Engineering
topic Software Engineering
Artificial Intelligence
url https://arxiv.org/abs/2510.13857