Intel / article

AgentChaos: Chaos Engineering for Agent Systems via Programmatic Fault Injection

Source: Gou Tan, Zhensu Sun, Jieke Shi, Ting Zhang, Zilong He, Qingfu Wu, Shuai Liang, Weifeng Sun, Junda He, Pengfei Chen, Chuanfu Zhang, Lwin Khin Shar, David Lo
Author: Gou Tan, Zhensu Sun, Jieke Shi, Ting Zhang, Zilong He, Qingfu Wu, Shuai Liang, Weifeng Sun, Junda He, Pengfei Chen, Chuanfu Zhang, Lwin Khin Shar, David Lo
Date: Published Aug 10, 2026

Terms in this piece · Glossary

AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.

Why it matters

Shows that robustness tracks system implementation rather than model choice, and gives a source-code-free way to test how your own agent handles a bad API response.

Read the source arxiv.org