AI and the Future of Conflict

Research on how AI systems are reshaping conflict dynamics—from military operations and strategic decision-making to diplomacy, mediation, and peacebuilding—alongside practical methods to evaluate these tools before they are trusted in high-stakes settings

Artificial intelligence, ranging from machine learning and large language models to agentic systems and autonomous platforms, is changing how power is generated and applied. It is not only altering tactics on the battlefield, but also the strategic calculus behind national decisions, alliance behavior, escalation management, and crisis response. Futures Lab research connects emerging AI capabilities to real operational and policy workflows, with an emphasis on what is changing, what is not, and what risks grow as AI becomes embedded in security institutions.

Futures Lab's research on AI is organized around three lines of effort:

Across militaries and governments, AI is reshaping operational advantage and institutional design. This line of effort examines how AI-enabled tools affect command and staff processes, force planning, information operations, and professional military education, alongside how AI-driven wargaming and decision support can shape how leaders perceive options, risks, and time. It also assesses how agentic AI may shift organizational structures in defense and foreign policy institutions, and how these shifts affect speed, accountability, and control in high-pressure environments.

Trust in AI starts with measurement. In national security and crisis settings, where “ground truth” is contested, data are sensitive, and outcomes depend on context—standard evaluation approaches are often insufficient. This line of effort advances benchmarking methods that stress-test AI systems for reliability, robustness, bias, and risk under realistic conditions (e.g., uncertainty, missing information, adversarial pressure, and degraded communications). It includes assessing whether models exhibit decision-relevant failure modes tied to escalation dynamics, such as overconfidence, misreading adversary intent, premature use-of-force recommendations, or instability under time pressure, before integration into planning or advisory workflows. The goal is to ensure AI systems increase performance and resilience rather than introduce new vulnerabilities.

Demand for mediation and peacebuilding is rising, while practitioners face persistent constraints: information overload, fragmented data, time pressure, and the need to maintain trust and confidentiality. This line of effort explores how AI can support mediators and diplomats with bounded, practical capabilities—organizing large document sets, mapping stakeholders and issues, tracking proposals across drafts, and generating scenarios that expand the negotiation space—without replacing human judgment or relationship-building. Emphasis is placed on secure use, careful deployment, and practitioner-informed evaluation to ensure tools are usable in real mediation contexts.

Contact Information

AI, War, and Strategy

AI Evaluation and Benchmarking

AI, Diplomacy, and Conflict Resolution

All AI and the Future of Conflict Content

Filter by

1 - 10 of 34 results

AI and Grand Strategy: The Case for Restraint

Artificial intelligence may not reward a military arms race, but instead favor strategic restraint, economic capacity, and the ability to mobilize capital, infrastructure, and partnerships that shape long-term power and influence.

Report by Erica D. Lonergan and Benjamin Jensen — January 28, 2026

The U.S. Army and a Second Manhattan Project for AI

To win the AI race, the U.S. Army must again serve as a national integrator—mobilizing talent, securing infrastructure, and modernizing forces, just as it did during the original Manhattan Project.

Commentary by Jake S. Kwon and Benjamin Jensen — November 21, 2025

Will Trump’s Peace Plan for Ukraine Succeed?

Trump’s peace plan for Ukraine could open the door to negotiations—but its territorial concessions, weak security guarantees, and amnesty provisions risk making the deal collapse before it begins.

Critical Questions by Benjamin Jensen and Yasir Atalan — November 21, 2025

What Could a Trump Peace Plan in Ukraine Look Like?

As Trump’s team pushes for a Ukraine peace deal amid new pressure on Russia, experts and AI models outline the most likely outcomes, from frozen frontlines to the economic and security terms shaping negotiations.

Critical Questions by Benjamin Jensen and Yasir Atalan — October 17, 2025

What Does Lethality Really Mean in Modern War?

What does “lethality” really mean in modern war? Beyond warrior ethos, it’s about cohesion, adaptation, and multidomain operations—where drones, AI, and resilient forces redefine combat power for the twenty-first century.

Critical Questions by Benjamin Jensen — October 2, 2025

Benchmarking as a Path to International AI Governance

AI benchmarking could serve as a practical foundation for international AI governance related to security issues, enabling confidence-building measures between states while advancing U.S. interests in global AI standardization

Commentary by Ian Reynolds — August 5, 2025

Toward Reliable AI, from the Bottom Up

The Trump administration’s AI Action Plan calls for reliable AI, but achieving this requires bottom-up-driven benchmarking that engages civil society, not just tech companies and government, to ensure AI performs safely in critical applications such as national security.

Commentary by Ian Reynolds — July 28, 2025

Deterrence Runs on Rare Earths

The future of deterrence won’t be decided by tanks or troops alone—but by tiny, rare earth elements that power AI, drones, and defense systems. The United States’ strategic edge depends on securing what’s buried beneath the surface.

Commentary by Jake Kwon and Benjamin Jensen — July 24, 2025