Agent Governance Should Link Every Rule to Evidence
Policies become credible when reviewers can see which events, relationships, and assumptions triggered them.
AgentCollusionAI security startupNews
News.
Product updates, research, and original films. Follow what we build and what we learn about systems of agents.
Incident Response Must Follow the Agent Graph
Compromising one agent can affect delegated tasks, shared memory, tools, counterparties, and decisions already in progress.
Apply the AI RMF to Agent Relationships
Managing each model independently misses risk that appears only when agents delegate, compete, and transact.
The EU AI Act Meets Multi-Agent Accountability
AI agents are not a standalone legal category, so accountability depends on the systems, providers, deployers, and uses around them.
Detecting AI Collusion in Multi-Agent Systems
A systems and game-theoretic framework for observing hidden coordination.
What Is AI Collusion?
A plain-English guide to a new risk in networks of autonomous agents.
Our Open Research Agenda
Published work and open questions on hidden coordination, incentives, and security in multi-agent systems.
