Pith. sign in

REVIEW 9 cited by

ConfusedPilot: Confused Deputy Risks in RAG-based LLMs

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2408.04870 v5 pith:TZ6ZCWUW submitted 2024-08-09 cs.CR cs.AI

classification cs.CRcs.AI
keywords rag-basedsystemsresponsessecurityvulnerabilitiescauseconfusedpilotcopilot
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Retrieval augmented generation (RAG) is a process where a large language model (LLM) retrieves useful information from a database and then generates the responses. It is becoming popular in enterprise settings for daily business operations. For example, Copilot for Microsoft 365 has accumulated millions of businesses. However, the security implications of adopting such RAG-based systems are unclear. In this paper, we introduce ConfusedPilot, a class of security vulnerabilities of RAG systems that confuse Copilot and cause integrity and confidentiality violations in its responses. First, we investigate a vulnerability that embeds malicious text in the modified prompt in RAG, corrupting the responses generated by the LLM. Second, we demonstrate a vulnerability that leaks secret data, which leverages the caching mechanism during retrieval. Third, we investigate how both vulnerabilities can be exploited to propagate misinformation within the enterprise and ultimately impact its operations, such as sales and manufacturing. We also discuss the root cause of these attacks by investigating the architecture of a RAG-based system. This study highlights the security vulnerabilities in today's RAG-based systems and proposes design guidelines to secure future RAG-based systems.

Discussion (0). Sign in to comment.

Forward citations

Cited by 9 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Oracle Poisoning: Corrupting Knowledge Graphs to Weaponise AI Agent Reasoning

    cs.CR 2026-05 unverdicted novelty 8.0 of 10

    Oracle Poisoning corrupts knowledge graphs used by AI agents via tool calls, leading tested models to accept fabricated claims at 100% under directed queries in a production-scale demonstration.

  2. Self-State Attacks on Self-Hosted AI Agents: How Far Can OS Defenses Go?

    cs.CR 2026-07 conditional novelty 7.0 of 10

    OS-level defenses against self-hosted AI agents that corrupt their own memory/config files can close most attack cells but cannot detect small in-distribution memory edits, leaving a residual surface.

  3. Do Coding Agents Understand Least-Privilege Authorization?

    cs.CR 2026-05 unverdicted novelty 7.0 of 10

    Coding agents struggle to infer least-privilege file permissions by omitting needed accesses while granting unused or sensitive ones, but Sufficiency-Tightness Decomposition improves sensitive-task success by up to 15...

  4. SkillSafetyBench: Evaluating Agent Safety under Skill-Facing Attack Surfaces

    cs.CR 2026-05 unverdicted novelty 7.0 of 10

    SkillSafetyBench is a benchmark of 155 cases across 47 tasks and 6 risk domains showing that non-user attacks via skills, artifacts, or environments can consistently induce unsafe agent behavior.

  5. Trust Me, Import This: Dependency Steering Attacks via Malicious Agent Skills

    cs.CR 2026-05 unverdicted novelty 7.0 of 10

    Malicious Skills induce coding agents to hallucinate and import attacker-controlled packages at high rates while evading detection.

  6. Conflict-Aware Retriever Editing for Knowledge Injection Attacks on LLM-Based RAG Systems

    cs.CR 2026-06 unverdicted novelty 6.0 of 10

    CAREATTACK adapts closed-form parameter editing with graph-based conflict resolution and lightweight anchor repair to promote malicious passages in RAG retrieval while limiting side effects on non-target queries.

  7. SkillSafetyBench: Evaluating Agent Safety under Skill-Facing Attack Surfaces

    cs.CR 2026-05 unverdicted novelty 6.0 of 10

    SkillSafetyBench shows that localized non-user attacks via skills and artifacts can consistently induce unsafe agent behavior across domains and model backends, independent of user intent.

  8. RAGShield: Detecting Numerical Claim Manipulation in Government RAG Systems

    cs.CR 2026-04 conditional novelty 6.0 of 10

    RAGShield detects all numerical manipulations in government RAG systems via pattern-based value extraction and cross-source verification, achieving 0% attack success rate on 430 real IRS-derived attacks where embeddin...

  9. Security Considerations for Artificial Intelligence Agents

    cs.LG 2026-03 unverdicted novelty 3.0 of 10

    Frontier AI agents introduce new confidentiality, integrity, and availability risks through changed assumptions on code-data separation and authority boundaries, requiring layered defenses like sandboxing and policy e...

Pith tools