Modeling Cooperation

Switzerland (distributed team)

https://www.modelingcooperation.com/

Actively Fundraisingearly7 peopleFounded 2019

Modeling Cooperation conducts research on the dynamics of competition for transformative AI development, using game theory and computational modeling to understand how competitive pressures affect AI safety outcomes. Founded in 2019 at the AI Safety Camp, the organization builds interactive research software tools that allow researchers and decision-makers to explore models of AI competition, and collaborates with AI governance academics on policy-relevant research. The team operates as a network of mission-aligned freelancers and volunteers, fiscally sponsored by Convergence Analysis.

Funding Details

Annual Budget: -
Monthly Burn Rate: -
Current Runway: -
Funding Goal: -
Funding Raised to Date: $329,316
Fiscal Sponsor: Convergence Analysis

Theory of Change

Modeling Cooperation believes that competition dynamics in AI development pose significant catastrophic risks, as competing teams may skimp on safety precautions to gain first-mover advantages. By using game theory and computational modeling to formally analyze these competitive dynamics, they aim to identify optimal strategies and policy proposals that could encourage safer competition. Their interactive software tools make these insights accessible to researchers and decision-makers, while their Intelligence Rising workshops give participants firsthand experience of how competitive pressures affect safety choices. The causal chain runs from rigorous analysis of competition models to policy-relevant insights to better-informed governance decisions that reduce the probability of catastrophic outcomes from AI development races.