We are a hybrid research institute, developing mathematical frameworks that address how AI systems transform collective decision-making dynamics and power structures.
We combine research in active inference, decentralized consensus mechanisms, and social choice theory to map intervention points that can shift inadequate equilibria toward more robust, sustainable systems that benefit humanity.
Who We Are
Equilibria Network is an interdisciplinary research organization focused on problems of destabilizing and emergent dynamics in complex system, and the problem of multi agent verifiable commit and trust mechanisms.
True to our interdisciplinary nature, we are a team of researchers from diverse backgrounds including complexity science, computational biology, systems theory and AI safety.
What We Do
Our work is founded on the premise that many AI risks stem from the interaction of AI with other complex systems. System level emergent risks need systems level emergent safety properties.
Identifying Core Needs in Technical AI Governance
Building a network of governance stakeholders to understand complex system challenges.
We work on building a network of technical AI governance stakeholders to understand on-ground real-world complex system challenges. Understanding the needs of research, industry, and policy helps us identify concrete problems where system level thinking is needed—ensuring our work addresses practical governance issues rather than theoretical abstractions.
Building Simulation Frameworks
Developing agent-based models to visualize policy outcomes and reveal cascading effects.
We build agent-based models that allow members of our network to visualize how their proposed policies might play out in practice. These simulations provide intuitive ways to test governance approaches before implementation, revealing potential cascading effects and unexpected macro system level consequences that might not be obvious from multi agent simulations alone.
Research Mathematical Foundations
Creating formal models of information flow, incentives, and coordination mechanisms.
To power these simulations accurately, we're developing formal mathematical models that capture how information flows through networks of agents, how incentives shape behavior across systems, and how different coordination mechanisms respond under pressure. This theoretical work underpins our simulations and ensures they reflect real-world dynamics accurately.
Work With Us
We are building a community of collaborators who share our fascination with collective intelligence and its role in shaping safer AI systems.
Researchers
Academic or independent researchers exploring collective intelligence
Learn morePractitioners
Those implementing AI governance or coordination mechanisms
Learn moreInterdisciplinary Connectors
Bridging traditionally separated fields and perspectives
Learn moreExperimentalists
Building and testing systems to study collective intelligence
Learn moreOur work thrives on diverse viewpoints and complementary skills. Whether you are a researcher, practitioner, advisor, or simply someone with a passionate interest in these questions, there are meaningful ways to contribute to our mission.
Get in Touch