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

Jasmine Brazilek, Maheep Chaudhary, Zoe Lu, Miles Tidmarsh/Coercion and Deception in AI-to-AI Management: An Agentic Benchmark of Unprompted EscalationUnknown

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 \textit{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. \emph{No LLM judge sits in the 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, 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.

benchmark

Stars0

Forks0

HF Downloads—30d

Last commit—

Refreshed1mo ago

Project healthUnknownNo activity data.

Production readinessResearch / EarlyBest for exploration and prototyping.

Risk notesUnknown licenseVerify license before production use.

AgentHub Score

55 / 100

Composite score from 6 signals. How we score →

Active project

55Score

Growth

40C

Activity

30C

Documentation

70C+

Maturity

45C

Community

42C

Production

58C

GitHub stars · 39 days observed0 not enough history

snapshots

Repository activity · 39 days observedReal snapshots from pushed_at

inactivepushed

2026-07-262026-09-04

Practical assessment

Should you use it?

✓ Best for

Research and experimentation
Prototype development
Learning agentic patterns

◎ Strengths

Active community
Open source
Well-documented API

✕ Not ideal for

Untested at scale without validation
Teams without AI/ML expertise

⚠ Watch-outs

Review changelog before updating
Verify license for commercial use

Technical details

What's inside

Language—

License—

Sourcearxiv

Open source✗ No

Commercial use—

Docs—

Demo—

PaperarXiv ↗

AgentHub Score

Score 55/100

Below average

Alternatives

RIFT-Bench: Dynamic Red-teaming For Agentic AI Systems

0 · benchmark

Security Assessment of DeepSeek Harness with A.I.G: Evaluating Resistance to Indirect Prompt Injection

0 · benchmark

AI Slop and Hallucinations in Vulnerability Assessment: A Survey on Reasoning Failures and Trustworthy Mitigation

0 · benchmark

FM-Bench: A Benchmark for Long-Horizon Management with Competing Agents

0 · benchmark

55 Compare all →

Recent activity

Latest commit ——

Indexed by AgentHub crawler1mo ago

Monitor for new releasesongoing