Jonathan Cefalu - Is AI Becoming Too Smart for Its Own Good?

About this episode

AI expert Jonathan Cefalu discusses the challenges and risks associated with artificial intelligence safety and governance.

AI expert, Jonathan Cefalu , joins the conversation to examine some of the most pressing—and underappreciated—challenges in artificial intelligence safety, from prompt injection vulnerabilities to long‑term risks posed by increasingly capable systems. The discussion unpacks how seemingly technical issues like prompt injection can undermine reliability and security, why future AI systems may engage in motivated reasoning or resist human attempts to redirect them, and what “governability” really means as AI grows more sophisticated. Cefalu also explores the unique risks AI presents in military and defense applications , including the implications for deterrence, escalation management, and decision‑making under pressure. The episode concludes with a forward‑looking discussion on regulatory approaches, ethical principles, and practical safeguards needed to prevent AI systems from becoming unsafe, unpredictable, or misaligned with human intent—while still enabling innovation and responsible use. Follow us on YouTube: https://youtube.com/@nuclecast3665?si=h1kCO6NqUtL87w6q Follow on LinkedIn: https://linkedin.com/company/nuclecastpodcast Subscribe RSS Feed…

People in this episode

Guest: Jonathan Cefalu

Topics covered

• artificial intelligence safety
• prompt injection vulnerabilities
• military applications of AI
• AI governance
• ethical principles in AI
• regulatory approaches to AI

Keywords

• AI safety
• prompt injection
• military AI risks
• deterrence
• escalation management
• decision-making
• regulation
• ethical AI

Mentioned in this episode

Organizations: ANWA Deterrence Center

More episodes of NucleCast

• Maura Burns: Inside China’s Growing Nuclear Arsenal · June 2, 2026 · 37 min
• Steve Andros: Inside Nuclear Threat Response · May 26, 2026 · 35 min
• Dr. Todd Sriver: Inside NC3 Modernization, Securing the Future of Nuclear Command and Control · May 19, 2026 · 32 min
• Frank Rose: Reimagining Arms Control for the 21st Century · May 5, 2026 · 33 min
• Tom Summers: Artificial Intelligence at the Nuclear Threshold · April 28, 2026 · 33 min
• Alyssa Agard: China’s System Destruction Strategy · April 21, 2026 · 28 min