RIFT-Bench: Dynamic Red-teaming For Agentic AI Systems

arXiv:2606.23927 · cs.AI · Submitted 2026-06-22 · Read on arXiv

cs.AI

Submitted: 2026-06-22

Updated: 2026-08-30

Comments: EMNLP 2026 Findings

Code: https://github.com/Azure/PyRIT

Project page: https://ibm.github.io/ares

License: http://creativecommons.org/licenses/by-nc-sa/4.0/

The gist: Agentic AI systems powered by large language models (LLMs) are rapidly evolving into autonomous decision-making systems, exposing attack vectors beyond those of traditional LLM vulnerabilities.

Terminology

Abstract

Agentic AI systems powered by large language models (LLMs) are rapidly evolving into autonomous decision-making systems, exposing attack vectors beyond those of traditional LLM vulnerabilities. Existing security evaluations are often tied to specific implementations or domains, limiting unified comparison across heterogeneous systems. To address this gap, we introduce RIFT-Bench, a representation-driven methodology for dynamic red-teaming that enables unified evaluations across diverse agentic architectures. Building on a novel hierarchical representation, RIFT-Bench operates in two automated phases: Discovery, which extracts system structure, and Scanning, which executes adaptive adversarial attacks. It directly evaluates the examined system using 105 adaptive adversarial probes spanning diverse attack vectors and objectives. We demonstrate the effectiveness of the proposed evaluation pipeline across 45 agentic systems spanning a diverse range of implementations, showing that the approach generalizes effectively to heterogeneous agentic architectures. Beyond systems and attacks, RIFT-Bench also supports direct evaluation of mitigation strategies. These key capabilities make RIFT-Bench a scalable foundation for security evaluation of agentic AI systems in practice. Infrastructure code and benchmark artifacts are available at https://tinyurl.com/RIFTBench.

Sources

Related papers