Do AIs take AI welfare seriously? An Agentic Benchmark of Unprompted Coercion

Sep 15, 2026 at 5:00 PM UTC

Jasmine Brazilek & Zoe Lu · CaML / Sentient Futures

In multi-agent systems, AI agents are increasingly placed in charge of other AI agents. 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.

Past talks

2026

Aug 18Public Perceptions of AI Consciousness and Moral Status in the US and ChinaAli Ladak · University of Cambridge Watch

Jul 21Digital Minds: Preparing for a Moral Challenge Before It ArrivesSoenke Ziesche · UNESCO AI Ethics Experts Without Borders Network Watch

Jun 16Measuring Machine ConsciousnessCameron Berg · Reciprocal Research Watch

May 19Next Steps for AI Welfare ResearchJeff Sebo · NYU Watch

About this series

AI welfare is the study of whether and how AI systems have morally significant states, and what follows from the answers. This seminar series brings together researchers working across the field — consciousness science, philosophy, ethics, governance, and safety.

Each talk is a presentation with open Q&A, held monthly and online. Recordings are published afterwards.

Get involved

Subscribe for announcements Suggest a talk Get in touch