cs.AIcs.MAeess.SYSurvey
Engineering Trustworthy Agentic AI for Critical Systems
Jul 20, 2026
This paper surveys the trustworthiness of agentic AI systems in engineering domains, organizing it around five dimensions: safety, robustness, transparency, accountability, and privacy.
The paper treats trustworthiness as a first-class engineering property for agentic AI systems and provides a comprehensive survey of methods for ensuring trustworthiness.
Before reading this…
Applications
To understand this paper, make sure you know these concepts first:
Agentic artificial intelligence systems, capable of autonomous perception, planning, tool use, and multi-step action, are increasingly proposed for critical engineering domains where decisions carry physical, operational, or economic consequences. This survey addresses a gap in current literature by treating trustworthiness, whether agentic behavior can be verified, audited, and trusted under the constraints that engineering practice actually requires, as a first-class engineering property, rather than evaluating agentic AI by task capability alone. The study adopts a trustworthiness model organized around five cross-cutting dimensions: safety and constraint satisfaction; robustness and reliability; transparency and interpretability; accountability and auditability; and privacy and security. This is mapped onto an agentic assurance workflow spanning perception through audit. Building on this foundation, agentic systems architectures, threats, concrete trust mechanisms, and quantitative metrics are surveyed for direct application in agentic systems development and evaluation. These principles are then examined across four constraint-bound engineering domains: power systems, autonomous vehicles/robotics/UAVs, high-performance computing, and communication networks, identifying recurring design patterns, shared failure modes, and domain-specific gaps. Synthesizing across those domains, agentic AI trustworthiness is shown to be a single problem, with a path outlined toward a reusable, cross-domain assurance framework analogous to the graded certification regimes used by mature safety-critical engineering fields.
LLM-Powered Agentic AI for 5G/6G Networks: A Tutorial and Survey on Architectures, Protocols, and St…
This paper presents a tutorial-and-survey on integrating agentic AI into Next-Ge…
Agents in the Wild: Where Research Meets Deployment
This tutorial explores advances and challenges in deploying large language model…
Early Adoption of Agentic Coding Tools by GitHub Projects
This paper analyzes 25,264 agentic pull requests from 2,361 GitHub repositories…
SolarChain-Eval: A Physics-Constrained Benchmark for Trustworthy Economic Agents in Decentralized En…
This paper proposes SolarChain-Eval, a physics-constrained benchmark for evaluat…
StructureClaw: Traceable LLM Agents and an Executable Benchmark for Structural Engineering Workflows
This paper introduces StructureClaw, an artifact-centered workbench for evaluati…
From Agentic to Autogenic Network Management for AI-Native 6G and Beyond: A Standards Perspective
This paper proposes Autogenic network management, a self-programming extension t…
Implicit Fine-tuning via Context Engineering: A Curriculum Learning Framework for Multimodal Entity…
This paper proposes PTFEA, a curriculum-learning-inspired framework that transla…
A Methodology for Auditable Trustworthiness Levels in AI Lifecycle Governance
The paper proposes a methodology for auditable trustworthiness levels in AI gove…