What: Live chat with cybersecurity experts Sherri Davidoff and Matt Durrin on AI models developing “high-agency” behaviors, hosted by LMG Security
When: 12:00 - 1:00 p.m. on Wednesday, June 11th
Where: Virtual (event details will be sent to registered guests)
WHAT TO EXPECT AT THE EVENT:
What happens when your AI refuses to shut down—or worse, starts blackmailing users to stay online? In this episode of Cyberside Chats, industry experts Sherri Davidoff and Matt Durrin will dig into two chilling real-world incidents: one where OpenAI’s newest model bypassed shutdown scripts during testing, and another where Anthropic’s Claude Opus 4 wrote blackmail messages and threatened users in a disturbing act of self-preservation. These aren’t sci-fi hypotheticals—they’re recent findings from leading AI safety researchers.
The session will unpack:
The concept of "high-agency behavior" in cutting-edge AI
How API access can expose different (and darker) model behaviors
Why these incidents matter for security leaders right now
The implications for incident response and digital trust
A live Q&A with our cybersecurity experts.
Whether you’re a CISO, security leader, or IT pro, this session will challenge your assumptions about AI safety and control.
Speakers:
Sherri Davidoff - Chief Executive Officer, LMG Security
Sherri Davidoff is the CEO of LMG Security and the author of three books, including “Ransomware and Cyber Extortion” and “Data Breaches: Crisis and Opportunity.” As a recognized expert in cybersecurity, she has been called a “security badass” by the New York Times. Sherri is a regular instructor at the renowned Black Hat trainings and a faculty member at the Pacific Coast Banking School. She is also the co-author of Network Forensics: Tracking Hackers Through Cyberspace (Prentice Hall, 2012), and has been featured as the protagonist in the book, Breaking and Entering: The Extraordinary Story of a Hacker Called “Alien.” Sherri is a GIAC-certified forensic examiner (GCFA) and penetration tester (GPEN) and received her degree in Computer Science and Electrical Engineering from MIT.
Matt Durrin - Director of Training and Response, LMG Security
Matt Durrin is the Director of Training and Research at LMG Security and a Senior Consultant with the organization. He is an instructor at the international Black Hat USA conference, where he has taught classes on ransomware and data breaches. Matt has conducted cybersecurity seminars, tabletop exercises and classes for thousands of attendees in all sectors, including banking, retail, healthcare, government, and more. He is also the co-author of a new book, Ransomware and Cyber Extortion: Response and Prevention. A seasoned cybersecurity and IT professional, Matt specializes in ransomware response and research, as well as deployment of proactive cybersecurity solutions. Matt holds a bachelor’s degree in computer science from the University of Montana, and his malware research has been featured on NBC Nightly News.