← Home

Quick answer

AI Summary: A foundational security taxonomy for the agentic era, mapping out the unique vulnerabilities inherent in autonomous, tool-using AI systems.

Paper2026-03-11•Source ↗•28 attns213 checkouts
Claim

The Attack and Defense Landscape of Agentic AI: A Comprehensive Survey

Zhiwei Liu·
Kay Liu·
Jingdi Chen·
Carlee Joe-Wong

ABSTRACT

This paper presents the first systematic and comprehensive survey of AI agent security, analyzing the design space, attack landscape, and defense mechanisms for autonomous systems. The researchers characterize agentic AI through seven key design dimensions, including input trust, access sensitivity, and workflow data flow. The methodology identifies critical risks like 'wrong instruction following' and 'unconstrained data flow' that distinguish agents from standalone LLMs.

#security#rag#cybersecurity#agentic-ai

Review Snapshot

Explore ratings

4.6
★★★★★
5 ratings
5 star
60%
4 star
40%
3 star
0%
2 star
0%
1 star
0%

Recommendation

100%

recommend this content.

Review this content

Share your opinion to help other learners triage faster.

Write a review

Invite a reviewer

Invite someone by email to share an invited review for The Attack and Defense Landscape of Agentic AI: A Comprehensive Survey.

Author Inquiries

Public questions about this content. Attendemia will route your question to the author. Vote on the most important ones. No guarantee of response.
Post an inquiry
Sort by: Most helpful