This report represents the ongoing policy debate on the race to artificial general intelligence (AGI) in a mathematically neutral model that allows policymakers to compare the outcomes of alternative strategies in international competition. The model suggests that incentives will continue to be aligned with accelerated development until coordination mechanisms are designed that are grounded in a common knowledge of the global risks of AGI.

A Prisoner’s Dilemma in the Race to Artificial General Intelligence

Lisa Abraham, Joshua Kavner, Alvin Moon

ResearchPublished Dec 1, 2025

The purpose of this report is to represent the ongoing policy debate on the race to artificial general intelligence (AGI) in a mathematically neutral model that allows policymakers to compare the outcome of alternative strategies in international technology competition. The analysis assumes that outcomes are driven by a strategic choice of whether to accelerate the development of AGI based on the trade-off between the perceived benefits of securing a first-mover advantage and the perceived risks, such as developing an uncontrolled or unaligned AGI, misuse by non-experts to create new weapons of mass destruction, instability between great powers that escalates to conflict, and other threats to human survival, flourishing, and global security. Although the model does not capture the full complexity of present-day racing dynamics, it suggests that incentives will continue to be aligned with accelerated development until coordination mechanisms are designed that are grounded in a common knowledge of the global risks of advancing toward AGI. This report is intended for policymakers and general audiences who are interested in understanding these racing dynamics and geopolitical implications.

Key Findings

If both the United States and China have the same shared assessment that the benefits of being the first to achieve AGI outweigh the risks, they are effectively locked in a prisoner’s dilemma

  • Each country is incentivized to accelerate the development of AGI, fearing that more cautious development will allow the other to gain a decisive advantage, reinforced by the logic of first-mover advantage.
  • As a result, both the United States and China will prioritize acceleration over risk mitigation.

If a common shared assessment emerges that the risks of catastrophic harm from accelerated development of AGI exceed the potential benefits of a first-mover advantage, the incentives can be aligned toward cooperation

  • In this context, both countries would benefit from coordinating their AGI development strategies to mitigate shared risks.
  • Aligned incentives for this type of cooperation will be implementable only if there are robust mechanisms for the United States and China to share information to establish common knowledge and verify mutual commitments to AGI development strategies.

Verification methods that allow countries to credibly signal that they are cooperating can increase the chance that cooperation can be sustained

  • These mechanisms are even more relevant in repeated-decision contexts; punishments can be effective only if countries have a way of detecting that their rivals are not cooperating.
  • In the absence of these mechanisms, mistrust is likely to persist, and the opportunity for genuine cooperation, especially when it is most vital, could be lost.

Topics

Document Details

Citation

Chicago Manual of Style

Abraham, Lisa, Joshua Kavner, and Alvin Moon, A Prisoner’s Dilemma in the Race to Artificial General Intelligence. Santa Monica, CA: RAND Corporation, 2026. https://www.rand.org/pubs/research_reports/RRA4245-1.html.
BibTeX RIS

Research conducted by

This research was independently initiated and conducted by the Center for the Geopolitics of Artificial General Intelligence with RAND Global and Emerging Risks using income from operations and gifts from RAND supporters, including philanthropic gifts.

This publication is part of the RAND research report series. Research reports present research findings and objective analysis that address the challenges facing the public and private sectors. All RAND research reports undergo rigorous peer review to ensure high standards for research quality and objectivity.

This document and trademark(s) contained herein are protected by law. This representation of RAND intellectual property is provided for noncommercial use only. Unauthorized posting of this publication online is prohibited; linking directly to this product page is encouraged. Permission is required from RAND to reproduce, or reuse in another form, any of its research documents for commercial purposes. For information on reprint and reuse permissions, please visit www.rand.org/pubs/permissions.

RAND is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.