In this after-action report, the authors describe a series of exercises centered on the misuse of artificial intelligence (AI) models to generate novel biological threats. These exercises illuminate critical decision points, capability gaps, and institutional challenges that may emerge as AI reshapes the strategic landscape—enabling policymakers to better prepare for an uncertain but potentially transformative technological future.

Infinite Potential—Insights from the Viral Uplift Scenario

After-Action Report from a Sequence of Day After Artificial General Intelligence Exercises

Anton Shenk, Matt Chessen, Barbara Del Castello, Gregory Smith, Richard S. Girven

ResearchPublished Apr 6, 2026

How should the United States prepare for and respond to potential artificial intelligence (AI) and artificial general intelligence (AGI) developments? The authors conducted a series of exercises centered on the misuse of AI models to generate novel biological threats. The ten exercises involved 119 participants drawn from senior government officials, analysts, and subject-matter experts and focused on a scenario involving accidental release of an AI-generated virus, leading to a global pandemic, as well as multiple state and nonstate actors racing to acquire similar AI-enabled biological weapons.

These exercises illuminate critical decision points, capability gaps, and institutional challenges that may emerge as AI reshapes the strategic landscape—enabling policymakers to better prepare for an uncertain but potentially transformative technological future. Among other observations, participants identified enhanced biological surveillance and early warning systems as the highest-priority capability need.

Key Findings

Participants’ assessments of the nature and severity of threats to U.S. national security and the policy responses the participants recommended consistently hinged on the judgments they made about key issues

  • The key issues include whether AI is a force multiplier or a game changer for biological threats, whether it is more fruitful to restrict AI models or to focus on specific threat actors, what the escalation and deterrence thresholds might be, and the tension between security and scientific and commercial progress.

The key issues suggest a set of capabilities and playbooks that could be developed to improve U.S. government preparedness for advanced AI

  • Among the proposed capabilities are proliferation monitoring and analysis, enhanced biological surveillance and early warning systems, and increased intelligence and counterintelligence.
  • The playbooks suggest ways to address the issues using the proposed capabilities.

Topics

Document Details

Citation

Chicago Manual of Style

Shenk, Anton, Matt Chessen, Barbara Del Castello, Gregory Smith, and Richard S. Girven, Infinite Potential—Insights from the Viral Uplift Scenario: After-Action Report from a Sequence of Day After Artificial General Intelligence Exercises. Santa Monica, CA: RAND Corporation, 2026. https://www.rand.org/pubs/research_reports/RRA4727-1.html.
BibTeX RIS

Research conducted by

This effort was independently initiated and conducted by the Center for the Geopolitics of Artificial General Intelligence within RAND Global and Emerging Risks using income from operations and gifts from RAND supporters.

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.