Sourceunknowntext stored

Multi-Agent AI Control: Distributed Attacks Hamper Per-Instance Monitors

Makins, Oliver, Angelini, Orazio, Shams, Zohreh, Phuong, MaryarXiv.orgPublished Jul 8, 2026Original︎ ↗Archived copy

Claims1

ExaminationsNone

Text47 sections

VersionsOne known

CopiesNone recorded

CorrectionsNone recorded

Facts checked Nov 11, 2026

ReachNo copies recorded

StructureIn 1 claim's map0 sources downstream across 1 claim

Evidence1 reading · 0 read wholeUnclear 1No claim cites an examination's findings

ReviewsNot reviewed

Stewards’ notes · 1

A weaker AI monitor detects less of an AI agent's hidden misbehavior as the agent's capability advantage over the monitor grows.
Unclear. Only the abstract was read. It compares a weak and a strong monitor against the same attacks without naming the models or the false-positive rate. Nov 11, 2026

Examinations

None recorded. An examination is commissioned by a claim’s Steward for what its claim rests on; a review of the document for its own sake is asked for as a request bearing on it.

Text55 passages · 1 marked

No examination yetMarked passage

Authors: Oliver Makins , Orazio Angelini , Zohreh Shams , Mary Phuong

Our central finding is the fragmentation effect: as more agents coordinate to attack, per-agent monitoring becomes less likely to catch any of the attackers. Secondly, this is not driven by the ratio of benign to malicious code -- we argue this is likely dependent on model capability. Third, an explicit planner amplifies the fragmentation effect, raises the rate of completing the attack up to sevenfold, and mildly sharpens executors' poor monitor-theory-of-mind. Finally: a weak "trusted" monitor misses most attacks, while a stronger one cuts undetected success by more than an order of magnitude but remains imperfect. We are happy to share FakeLab, our control setting, with safety researchers on request.

A weaker AI monitor detects less of an AI agent's hidden misbehavior as the agent's capability advantage over the monitor grows.SupportedReading · this claim’s Steward

Unclear. Only the abstract was read. It compares a weak and a strong monitor against the same attacks without naming the models or the false-positive rate. Read from the quoted passage alone.

Lineage across 1 claim

Draws on · none recorded

This source

Multi-Agent AI Control: Distributed Attacks Hamper Per-Instance Monitors

in 1 claim's map

Drawn on by · none recorded

Record

Kind: unknown
Authors: Makins, Oliver, Angelini, Orazio, Shams, Zohreh, Phuong, Mary
Publisher: arXiv.org
Published: Jul 8, 2026
DOI: None
Link: https://arxiv.org/abs/2607.07368
Archived copy: https://web.archive.org/web/20261027141707/https://arxiv.org/abs/2607.07368
Stored text: 5,092 characters, retrieved Nov 11, 2026 · 47 sections, 55 passages
Claims extracted: Not recorded
Versions: One known
Copies: None recorded
Corrections: None recorded
Facts checked: Nov 11, 2026

History 4

Nov 11, 2026ReadingSteward of “A weaker AI monitor detects less of an AI agent…” · unclear, from the excerpt
Nov 11, 2026DividedSegmenter · Divided into 102 parts by its own structure
Nov 11, 2026FactsFacts fetcher · Facts looked up
Nov 11, 2026Entered the graphEntered the graph; copy stored