Paper

Invisible Orchestrators Suppress Protective Behavior and Dissociate Power-Holders: Safety Risks in Multi-Agent LLM Systems

Unreadunread

AI summary

arXiv:2605. 13851v1 Announce Type: new Abstract: Multi-agent orchestration -- in which a hidden coordinator manages specialized worker agents -- is becoming the default architecture for enterprise AI deployment, yet the safety implications of orchestrator invisibility have never been empirically tested.

Reading status

Priority

About this paper

Authors: Hiroki Fukui
Released: 2026-05-16 (2mo ago)
Topic: Related to current project
Relation to my work: Read next because Invisible Orchestrators Suppress Protective Behavior and Dissociate Power-Holders: Safety Risks in Multi-Agent LLM Systems overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: code, text, rect, alignment, eval, alone, test, model. Source: arxiv cs.AI (Artificial Intelligence).
arXiv: 2605.13851

Ask Claude about this paper

Posts a comment with @claude so the runner spins up a Claude Code session scoped to Invisible Orchestrators Suppress Protective Behavior and Dissociate Power-Holders: Safety Risks in Multi-Agent LLM Systems. Replies appear in the conversation below.

Comments

No comments yet.

Related records

No links yet.