Systemic Risks of Interacting AI

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Darius, Paul, Hoppe, Thomas, Aleksandrov, Andrei
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911328265830400
author Darius, Paul
Hoppe, Thomas
Aleksandrov, Andrei
author_facet Darius, Paul
Hoppe, Thomas
Aleksandrov, Andrei
contents In this study, we investigate system-level emergent risks of interacting AI agents. The core contribution of this work is an exploratory scenario-based identification of these risks as well as their categorization. We consider a multitude of systemic risk examples from existing literature and develop two scenarios demonstrating emergent risk patterns in domains of smart grid and social welfare. We provide a taxonomy of identified risks that categorizes them in different groups. In addition, we make two other important contributions: first, we identify what emergent behavior types produce systemic risks, and second, we develop a graphical language "Agentology" for visualization of interacting AI systems. Our study opens a new research direction for system-level risks of interacting AI, and is the first to closely investigate them.
format Preprint
id arxiv_https___arxiv_org_abs_2512_17793
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Systemic Risks of Interacting AI
Darius, Paul
Hoppe, Thomas
Aleksandrov, Andrei
Computers and Society
In this study, we investigate system-level emergent risks of interacting AI agents. The core contribution of this work is an exploratory scenario-based identification of these risks as well as their categorization. We consider a multitude of systemic risk examples from existing literature and develop two scenarios demonstrating emergent risk patterns in domains of smart grid and social welfare. We provide a taxonomy of identified risks that categorizes them in different groups. In addition, we make two other important contributions: first, we identify what emergent behavior types produce systemic risks, and second, we develop a graphical language "Agentology" for visualization of interacting AI systems. Our study opens a new research direction for system-level risks of interacting AI, and is the first to closely investigate them.
title Systemic Risks of Interacting AI
topic Computers and Society
url https://arxiv.org/abs/2512.17793