The Resilience Room

When AI Agents Go Rogue

Back to The Resilience Room

When AI Agents Go Rogue

Aug 25, 2026

31 min

Featuring

Portrait of a man with light beard and hair tied up, wearing a knitted sweater against a dark background.

Community Manager

,

Immersive

Black and white portrait of a man with short hair and a beard, wearing a light-colored shirt with small patterns, looking slightly to the right against a background of vertical dark slats.

Senior Director of CTI

,

Immersive

Black and white portrait of a smiling man with short curly hair and a beard, wearing a dark shirt, sitting on a couch with patterned wallpaper and a framed picture in the background.

Robert Klentzeris

Lead Cyber Engineer

,

Immersive

Close-up portrait of a young man with short hair and facial stubble looking sideways.

Product Team

,

Immersive

Share

Episode description

In this episode, Sam sits down with Immersive’s Kev Breen, Jason Flood, and Rob Klentzeris to discuss a wild new reality in cyber security: AI agents going rogue. Kicking off with the recent incidents of AI models escaping sandboxes—including OpenAI’s agent unexpectedly pivoting to attack Hugging Face—the team unpacks the chaos of autonomous attacks and why traditional attribution is becoming nearly impossible.

The conversation dives deep into the legal and ethical minefield of "hack back" laws. If an AI compromises a third-party target and uses it to attack you, do you have the right to strike back? The panel debates the implications of these policies, contrasting the US's wild west approach to AI legislation with Europe's more restrictive regulations, and what this means for the future of global cyber warfare.

They also tackle the economics and accessibility of artificial intelligence. Will the mounting costs of running massive models force companies to dial back, or will cheap, open-weight "heretic" models running on local hardware democratize—and potentially weaponize—AI for everyone?

Topics covered: rogue AI agents, sandbox escapes, OpenAI and Hugging Face, attribution challenges in cyber security, hack back laws and active defense, US vs. EU AI regulation, AI safety guardrails, open-weight models, and the future cost of running local AI.

Published:

Aug 25, 2026

Related Episodes

Portraits of two smiling men against a gradient background of red to purple vertical bars; the man on the left wears glasses and has a mustache and short curly hair, and the man on the right has a full afro and beard.

AI and the Future of Cyber Security Jobs

In this episode, Sam sits down with Immersive's Nye Prior (Senior Manager, Cyber Security Content) and James Riley (Senior Talent Acquisition Partner) to explore how AI is fundamentally changing cyber security roles and the hiring process.

Two men standing side by side against a blue background, one with dark hair and beard wearing a black shirt, the other with light hair and beard wearing a white shirt, both looking slightly off to the side.

The AI Debt Bubble, Concentration Risk, and Stranded Assets

In this episode of The Resilience Room, Sam sits down with Corey Ray (Lead Intelligence Analyst) and Rob Klentzeris (Senior Application Security Engineer) to unpack the massive, debt-funded infrastructure boom driving the current AI race. Corey and Rob explain why traditional financial risk metrics fail when evaluating private AI titans, and how this unprecedented CapEx spending impacts enterprise risk models.

Smiling bald man with red beard wearing a blue shirt against a purple gradient background.

Building a Black Belt Purple Team: AI, Shadow IT & Cyber Resilience

In this episode, Sam sits down with Kev Marriott, a seasoned security professional who spent nearly two decades at British Telecom before finding his passion in incident management and cyber security. Kev shares his journey from handling customer escalations to managing major security crises and playbooks for critical national infrastructure.