Making AI Possible

Making AI Possible - The Security Episode - November 2025

Episode Summary

In our November 3 session of Making AI Possible, Director of AI Programs, Nicholas Beaudoin, sits down with Ismael Montañez, Cyber Security Engineer, as they explore one of the most urgent and least understood frontiers of AI: security. We'll examine how LLM systems and the infrastructure that supports them introduce new layers of vulnerability, from data poisoning and prompt injection to model manipulation and adversarial exploits.

Episode Notes

We'll talk about the questions that every leader should be asking right now:

• Hacker vs. Cybercriminal: When AI joins the fight, do we face a new category? An AI Cyberhacker, or something even more autonomous?
• Securing the Foundations of AI: Why data integrity and trust are the real first lines of defense for both traditional and LLM-based systems.
• The Hidden Infrastructure: Protecting AI’s compute backbone from invisible threats.
• Adversarial AI: The rise of offensive machine learning, and what happens when the model becomes the weapon.
• The Human Factor: Building a security-first AI culture that does not rely on a few experts, but everyone.
• The future of responsible AI depends on one thing: treating cybersecurity not as a barrier, but as an enabler

Ismael Montañez “Ish” is a Cybersecurity Engineer, Ethical Hacker, and Purple Team Specialist with over six years of experience protecting AI-driven systems and cloud environments. He holds two bachelor’s degrees in Cyber Forensics with Network Security and Digital Media, blending technical precision with creative problem solving to design resilient security architectures. Before entering the private sector, Ish served as a Military Intelligence Sergeant, managing classified intelligence operations and developing expertise in threat analysis and secure information handling. He also spent four years in the financial industry, conducting IT and Linux forensics to safeguard critical data and digital assets. Ismael currently works at a leading artificial intelligence company, where he focuses on cybersecurity engineering, cloud security and engineering, AI security and oversight, and proactive defense strategies. His mission is to bridge the gap between human ingenuity and machine intelligence ensuring that innovation in AI remains secure, ethical, and resilient. ♦

As Caltech CTME's Affiliate Director of AI Programs, Nicholas Beaudoin intersects strategic program development with hands-on curriculum design, positioning AI/ML education programs at the forefront of technological transformation. Leading the AI programs at Caltech's Center for Technology & Management Education, Nicholas' focus is on delivering impactful learning experiences tailored to a diverse clientele, including private and public sectors, as well as government and non-profit organizations.  ♦