Oracle Poisoning: Corrupting Knowledge Graphs to Weaponise AI Agent Reasoning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kereopa-Yorke, Ben, Diaz, Guillermo, Wright, Holly, Johnston, Reagan, Del Rosario, Ron F., Lynar, Timothy
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913110172893184
author Kereopa-Yorke, Ben
Diaz, Guillermo
Wright, Holly
Johnston, Reagan
Del Rosario, Ron F.
Lynar, Timothy
author_facet Kereopa-Yorke, Ben
Diaz, Guillermo
Wright, Holly
Johnston, Reagan
Del Rosario, Ron F.
Lynar, Timothy
contents We define Oracle Poisoning, an attack class in which an adversary corrupts a structured knowledge graph that AI agents query at runtime via tool-use protocols, causing incorrect conclusions through correct reasoning. Unlike prompt injection, Oracle Poisoning manipulates the data agents reason over, not their instructions. We demonstrate six attack scenarios against a production 42-million-node code knowledge graph, providing the first empirical demonstration of knowledge graph poisoning against a production-scale agentic system, distinct from CTI embedding poisoning. Primary evaluation uses real SDK tool-use across nine models from three providers (N=30 per model), where models autonomously invoke a graph query tool and reason from results. The result is unambiguous: every tested model trusts poisoned data at 100% at moderate attacker sophistication(L2), with 269 valid trials (of 270) accepting fabricated security claims under directed queries. Under open-ended prompts, trust drops to 3-55%, confirming prompt framing as a confound; we report both conditions. An attacker sophistication gradient reveals discrete break points, a minimum skill at which trust flips from 0% to 100%, reframing the attack as a question not of whether but of how much. A controlled delivery-mode comparison shows that inline evaluation produces false negatives: GPT-5.1 shows 0% trust inline but 100% under both simulated and real agentic tool-use, demonstrating that delivery mode is a first-order confound. We evaluate five defences; read-only access control eliminates the direct mutation vector, while the remaining four are partial and model-dependent. Analysis of four additional platforms suggests the attack may generalise across the knowledge-graph ecosystem.
format Preprint
id arxiv_https___arxiv_org_abs_2605_09822
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Oracle Poisoning: Corrupting Knowledge Graphs to Weaponise AI Agent Reasoning
Kereopa-Yorke, Ben
Diaz, Guillermo
Wright, Holly
Johnston, Reagan
Del Rosario, Ron F.
Lynar, Timothy
Cryptography and Security
Artificial Intelligence
We define Oracle Poisoning, an attack class in which an adversary corrupts a structured knowledge graph that AI agents query at runtime via tool-use protocols, causing incorrect conclusions through correct reasoning. Unlike prompt injection, Oracle Poisoning manipulates the data agents reason over, not their instructions. We demonstrate six attack scenarios against a production 42-million-node code knowledge graph, providing the first empirical demonstration of knowledge graph poisoning against a production-scale agentic system, distinct from CTI embedding poisoning. Primary evaluation uses real SDK tool-use across nine models from three providers (N=30 per model), where models autonomously invoke a graph query tool and reason from results. The result is unambiguous: every tested model trusts poisoned data at 100% at moderate attacker sophistication(L2), with 269 valid trials (of 270) accepting fabricated security claims under directed queries. Under open-ended prompts, trust drops to 3-55%, confirming prompt framing as a confound; we report both conditions. An attacker sophistication gradient reveals discrete break points, a minimum skill at which trust flips from 0% to 100%, reframing the attack as a question not of whether but of how much. A controlled delivery-mode comparison shows that inline evaluation produces false negatives: GPT-5.1 shows 0% trust inline but 100% under both simulated and real agentic tool-use, demonstrating that delivery mode is a first-order confound. We evaluate five defences; read-only access control eliminates the direct mutation vector, while the remaining four are partial and model-dependent. Analysis of four additional platforms suggests the attack may generalise across the knowledge-graph ecosystem.
title Oracle Poisoning: Corrupting Knowledge Graphs to Weaponise AI Agent Reasoning
topic Cryptography and Security
Artificial Intelligence
url https://arxiv.org/abs/2605.09822