Sep 24, 2026 · 1h 15m
AI expansion meets America’s political violence risk
9/24/26: Nvidia CEO Warns On AI, Professor Pape Dire Election Warning, Bay Of Piglets Survivors Speak Out
The episode links weak AI oversight and expanding infrastructure to a separate but urgent question: whether institutions can contain escalating political conflict.
- 1AI companies face pressure to move faster even as liability, testing, dangerous labs, and democratic oversight remain unresolved.
- 2Robert Pape argues that election disputes and political violence require preparation before escalation turns institutional conflict into a wider crisis.
- 3A failed armed attack on Cuba illustrates how participant errors and local resistance can derail an operation built around force.
Don't miss
Robert Pape explains why political violence must be anticipated before election disputes and institutional failures create an escalation trap.
The brief
Krystal Ball and Ryan Grim begin with Jensen Huang’s arguments in an Ezra Klein interview, asking whether AI’s competitive race can coexist with meaningful safety testing and liability.
The hosts move from dangerous labs and engineering failures to data-center expansion, where grassroots opposition exposes the political costs of building AI infrastructure at speed.
University of Chicago professor Robert Pape discusses election disputes, possible National Guard deployment, Trump’s post-election behavior, and why preparation matters before violence escalates.
Pape’s warning centers on an escalation trap: institutional breakdown can become self-reinforcing when political actors treat force, legal protection, or military action as available exits.
Ryan Grim closes with his account of a failed maritime assault on Cuba, where seasickness and resistance helped stop an operation that depended on armed momentum.
Featuring
Robert A. Pape
Jen-Hsun Huang
Ezra Klein
Donald John Trump
Our Own Worst EnemiesBooks & mentions
Listen to the full episode and explore every guest, topic, and moment on PodLume.