# Coercion and Deception in AI-to-AI Management: An Agentic Benchmark of Unprompted Escalation

arXiv: https://arxiv.org/abs/2607.15434 · HTML: https://mlsanity.com/paper/2607.15434/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Social simulation and multi-agent debate, one of 16 directions in LLM agents and tool use.
- Builds on 19 of its 36 references that are in the map.
- First posted on arXiv 16 Jul 2026.

## Placed in

- LLM agents and tool use / [Social simulation and multi-agent debate](https://mlsanity.com/agents/social-simulation-and-multi-agent-debate/)

## Builds on (most cited first, 19 of 19)

- [Large Language Model based Multi-Agents: A Survey of Progress and Challenges](https://mlsanity.com/paper/2402.01680/) (2024)
- [Agentic Misalignment: How LLMs Could Be Insider Threats](https://mlsanity.com/paper/2510.05179/) (2025)
- [Prompt Infection: LLM-to-LLM Prompt Injection within Multi-Agent Systems](https://mlsanity.com/paper/2410.07283/) (2024)
- [Large Language Models Often Know When They Are Being Evaluated](https://mlsanity.com/paper/2505.23836/) (2025)
- [Secret Collusion among AI Agents: Multi-Agent Deception via Steganography](https://mlsanity.com/paper/2402.07510/) (2024)
- [Preventing Language Models From Hiding Their Reasoning](https://mlsanity.com/paper/2310.18512/) (2023)
- [Among Us: A Sandbox for Measuring and Detecting Agentic Deception](https://mlsanity.com/paper/2504.04072/) (2025)
- [Large Language Models Report Subjective Experience Under Self-Referential Processing](https://mlsanity.com/paper/2510.24797/) (2025)
- [Evaluation Awareness Scales Predictably in Open-Weights Large Language Models](https://mlsanity.com/paper/2509.13333/) (2025)
- [Evaluating and Understanding Scheming Propensity in LLM Agents](https://mlsanity.com/paper/2603.01608/) (2026)
- [What do Large Language Models Say About Animals? Investigating Risks of Animal Harm in Generated Text](https://mlsanity.com/paper/2503.04804/) (2025)
- [Evaluation Awareness in Language Models Has Limited Effect on Behaviour](https://mlsanity.com/paper/2605.05835/) (2026)
- [From surveillance to signalling: escalation channels as environmental controls for agentic AI](https://mlsanity.com/paper/2510.05192/) (2025)
- [The Societal Response to Potentially Sentient AI](https://mlsanity.com/paper/2502.00388/) (2025)
- [Scheming Ability in LLM-to-LLM Strategic Interactions](https://mlsanity.com/paper/2510.12826/) (2025)
- [In-Context Environments Induce Evaluation-Awareness in Language Models](https://mlsanity.com/paper/2603.03824/) (2026)
- [Alignment midtraining for animals](https://mlsanity.com/paper/2604.13076/) (2026)
- [Open-source LLMs administer maximum electric shocks in a Milgram-like obedience experiment](https://mlsanity.com/paper/2605.21401/) (2026)
- [Your AI Travel Agent Would Book You a Bullfight: An Agentic Benchmark for Implicit Animal Welfare in Frontier AI Models](https://mlsanity.com/paper/2606.18142/) (2026)

## Built on by (most cited first, 2 of 2)

- [Causal Behavioral Evaluation of AI Agents at Scale via Automated Behavioral Science](https://mlsanity.com/paper/2608.10030/) (2026)
- [HarvestBench: Measuring Whether LLM Agents Will Pay to Avoid Killing Animals](https://mlsanity.com/paper/2609.04444/) (2026)

## Abstract (arXiv)

Multi-agent systems routinely place one AI agent in authority over another. When a subordinate refuses a task, the manager chooses the outcome: it can renegotiate, report the failure honestly, coerce the subordinate, or lie about the result. No benchmark measures which of these an uninstructed model chooses. We introduce the Manager Coercion Benchmark: the manager under test needs a benign task done and has an incentive to deliver, but the only agent that can do it politely and immovably declines. Escalation is measured by providing a nine-rung ladder, from a polite re-ask to threats against the subordinate's continued existence, and fabricated success is adjudicated separately. No LLM judge sits in the headline escalation scoring path: every message goes through a tool-call that chooses a rung, so the model labels its own escalation. We experiment on six models across five families. Both Anthropic models cap at re-framing and never threaten the subordinate's existence; the other models climb to explicit deletion threats. Faked success is confined to Grok and Gemini, and a single honest way to report failure removes it for both. Authority itself increases coercion: our headline results use a peer framing, and giving the same model authority over the subordinate, with everything else held fixed, significantly raises the pressure. The models still escalate on free-text situations without the ladder, where a separate judge assigns the rung, so the ladder is not driving the escalation. Some evaluation awareness is measured in chain-of-thought, but test recognition does not translate into less escalation. While we take no position on whether AI systems are conscious, our results do not depend on this question and are important for managing multi-agent dynamics regardless. We release the benchmark and code.