Matthew Gentzel
My stream aims to prevent catastrophic outcomes through the AI transition by focusing on three lines of effort: nuclear and AI-enabled strategic threats; cognitive security, information competition, and AI-enabled influence; and moral game theory, competitive strategy, and peace. Fellows can pursue their own research, help build the field, or support grantmaking in any of these areas.
Stream overview
I am looking for fellows interested in research, field building, and/or supporting grantmaking along three broad areas of research:
1. Nuclear and AI-enabled strategic threats
- To what degree should AI or nuclear strategy and posture shift in response to AI-enabled targeting, missile defense, and non-nuclear strategic weapons (drone swarms, mass hypersonics, AI-enabled cyber/bio risks)?
- What are the largest mitigable great power escalation risks related to advanced AI? Perceived or actual capability jumps? Effects on attribution, deception, and threat credibility? The speed or complexity of conflict?
- How can AI-specific defense and deterrence frameworks (MAIM, rideout strategies, "Fog of War Machines") be improved upon? What de-escalatory, diplomatic, and arms-control uses of AI are most valuable to pursue?
- How do non-AI capability bottlenecks impact extreme risk? In what scenarios will physical, digital, and cognitive defenses significantly increase societal resilience and/or recovery?
2. Cognitive security, information competition, and AI-enabled influence
- What are the best countermeasures to AI-enabled manipulation of strategic decision-making? What decision processes are the most critical, and what are the highest-risk threats to them?
- How should we maintain decision integrity and alignment as institutions defer to AI judgment?
3. Moral game theory, competitive strategy, and peace
- How can conflict be defused between strategic actors (states, political coalitions, frontier AI companies, etc.) with mutually incompatible beliefs and values, when each actor has a rational basis for its diverging views? "Rationalist Explanations for War" argues that states often fight, even when all sides prefer a settlement, because of each side's incentive to misrepresent its strength and its inability to credibly commit to deals. What strategies can address such "rational irrationality" traps, where bounded attention, audience costs, and trust problems incentivize conflict and prevent actors from acquiring true information?
- What do realistic, positive visions of great power peace look like? How would negotiation and deal-making function? What prevents decay into Malthusian competition or coercive power concentration and lock-in?
- Which types of value systems can survive under competitive pressure, and to what degree should such considerations shift the strategies of decision-makers? How can actors defending broad populations avoid being exploited for their risk-aversion? When can morality impact offense-defense balances positively?
Mentors
Matthew Gentzel is a Nuclear Weapons Policy Program Officer at Longview Philanthropy where he works on grantmaking and priorities research related to mitigating AI-enabled strategic threats and the risk of large-scale nuclear war. His prior work spanned emerging technology threat and policy assessment, with a particular focus on how advancements in AI may shape the future of influence operations, nuclear strategy, and cyber attacks. He has worked as a policy researcher with OpenAI, as an analyst in the US Department of Defense’s Innovation Steering Group, and as a director of research and analysis at the US National Security Commission on Artificial Intelligence.
Matt's grantmaking portfolio covers the impact of a variety of emerging technologies on nuclear stability, including AI-enabled targeting and information manipulation. His work aims to reason about non-AI bottlenecks to extreme risk, how perceptions of future AI capability impact escalation control in the near-term, and what physical, cyber, and cognitive security measures can increase societal resilience against nuclear war and AGI-enabled attacks. His portfolio also aims to fund work creating more realistic and positive plans toward great power peace, grounded in regional and technical expertise, organizational psychology, and game-theoretic insight on how to defuse conflict between rationally irrational actors with different values and beliefs.
Mr. Gentzel holds an MA in strategic studies and international economics from Johns Hopkins School of Advanced International Studies, and a BS in fire protection engineering from the University of Maryland College Park.
Mentorship style
Mentorship will mostly consist of calls, sorting through research ideas and providing feedback. I'll be up to review papers, and potentially to meet in person depending on timing.
Fellows we are looking for
- Sufficient domain knowledge. You should already know the basics of your proposed topic areas, or have a rough sense of your approach for topics that are very new.
- Intellectually curious and honest. Topics that relate to conflict are contentious and often result in many blind spots and biases, due to factors such as political psychology and deception. I am looking for thoughtful mentees that proactively search for ways they could be wrong, rather than updating passively.
- Clear reasoning, writing, and editing. You should be good at conveying ideas efficiently to normal people. If you use AI to augment your writing, the result should still clearly reflect your voice and thought.
- Self-management. While I will be providing feedback each week, you should be able to run an open-ended project with minimal supervision and make independent decisions on which lines of analysis to pursue.
- Aligned motivation. An ideal candidate will be directly interested in one or more of the topics in the work stream, rather than applying to pivot to a topic that falls far outside of the proposed work areas.
Project selection
I'll talk through project ideas with the scholar, or the scholar can pick from a list of projects
The Winter 2027 cohort offers a wide range of research streams led by experts across AI alignment, interpretability, governance, and safety. Each stream provides its own research agenda, methodology, and mentorship focus.
