arxiv
PublishedJuly 23, 2026 at 4:00 AM
—neutral
Coercion and Deception in AI-to-AI Management: An Agentic Benchmark of Unprompted Escalation
Publisher summary· verbatim
arXiv:2607.15434v3 Announce Type: replace-cross Abstract: 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.
Stay posted· Newsletter
A 5-min weekly brief — top movers, price watch, story of the week.
Discussion
No replies yet. Be first.
Related coverage
More from ARXIV
arxivBringing Value Models Back: Generative Critics for Value Modeling in LLM Reinforcement Learning4harxivSubagents vs Agent Skills: Executing Reusable Knowledge for Long-Horizon Agentic Tasks4harxivDistribution-Consistent Inference for Dynamic Sparse Mixture-of-Experts4harxivIn RAG We Trust? Measuring Robustness of Retrieval-Augmented Generation Under Document Poisoning4hThe Bubble Brief
WEEKLYRead multiagent insights every Tuesday — top movers, new releases, story of the week.
Originally published on arxiv ↗