The Real Risks of LLMs - Guardrails, Judgment, and the Human Element in Cybersecurity

In this episode of Data Driven, hosts Frank La Vigne, Candace Gillhoolley, and BAILeY sit down with Mike Armistead, CEO of Pulse Security AI—a cybersecurity veteran who's been fortifying digital defenses since before AI made headlines and hackers had professional profiles. Together, they dig into the dual-edged sword that is artificial intelligence in cybersecurity, exploring how AI serves as both a powerful tool against cyber threats and a potential weapon in the wrong hands. Mike Armistead shares stories from the front lines, including his experience during the "code red" era at Google when ChatGPT shook up the tech world, and offers real-world advice on why LLMs (large language models) aren’t the magic fix for every problem—and why they desperately need guardrails. You’ll hear why your next big data breach could be hiding in a cleverly crafted AI prompt, why humans still matter when it comes to judgment calls, and why good old-fashioned security hygiene is as critical as ever. Whether you’re a developer, data scientist, or just password-paranoid, this episode will make you rethink how you approach security in the age of AI. Tune in for expert insights, hard-earned lessons, and a few laughs as the Data Driven crew uncovers where technology, risk, and "common sense" collide. Time Stamps 00:00 AI-Assisted Cybersecurity for SOCs 04:26 "AI Rush and LLM Insights" 09:12 AI-Powered Cybersecurity Strategy Insights 10:01 "Cybersecurity, ChatGPT, and Impressions" 13:17 AI Tools: Power and Risks 18:06 "Teaching Critical Thinking in AI Era" 20:59 "Guardrails and Next-Gen AI Systems" 24:22 Human Judgment vs AI Limitations 27:37 "Pressure Testing for Accuracy" 30:09 Future Tech Advancements and Challenges 34:58 "Risk Awareness Beyond Compliance" 37:38 "Cybersecurity Risks and AI Defense" 41:54 Cybersecurity Risks and Preparedness 43:04 "Situational Security in Practice" 46:05 "Cybersecurity's Evolving Threat Landscape" 51:52 "Builders vs. Destroyers Mindset" 55:05 Modern Password Practices 56:39 "Pulse Security AI & Community"

You can listen to The Real Risks of LLMs - Guardrails, Judgment, and the Human Element in Cybersecurity online on Radio and Podcast. Open the player on this page to stream the available audio.

The Real Risks of LLMs - Guardrails, Judgment, and the Human Element in Cybersecurity is an episode from Data Driven by Data Driven.

This episode is 58:02 long.

This episode was published on Nov 26, 2025.

Yes. Use the heart button on the episode page to add it to your favorite episodes list.

Yes. This page shows related episodes from Data Driven when more episodes are available from the podcast feed.