AI Unlocked: The Prompt Hacking Threat Landscape
This podcast explores the emerging threats of prompt hacking and the adversarial misuse of AI by analyzing real-world examples and security guidelines for Large Language Models.
🎧 Listen to this Episode
Show Notes
Delve into the critical security vulnerabilities of Artificial Intelligence, exploring the dangerous world of prompt injection, leaking, and jailbreaking as highlighted in SANS' Critical AI Security Controls and real-world adversarial misuse of generative AI like Gemini by government-backed actors. Understand how malicious actors attempt to bypass safety controls, extract sensitive information and manipulate LLMs for nefarious purposes, drawing insights from documented cases involving Iranian, PRC, North Korean, and Russian threat actors. Learn about the offensive techniques used and the ongoing challenge of securing AI systems,
Share this episode
Enjoying CISO Insights?
Subscribe to get new episodes delivered directly to your podcast app.
Related Episodes
Dismantling the FIMI House of Cards: The Fight for Information Integrity
This episode explores how the European Union is deploying a proactive deterrence strategy to dismantle the AI-enhanced information manipulation supply chains orchestrated by state threat actors...
▶️ Listen Now
The 90% Attack: Inside the First AI-Orchestrated Cyber Espionage Campaign
The disclosure of Chinese state-sponsored hackers using Anthropic’s Claude Code to autonomously execute 80–90% of a large-scale cyber espionage campaign confirms that AI has fundamentally altered the ...
▶️ Listen Now
The Algorithmic Adversary: Tracking the Shift to Novel AI-Enabled Malware
Threat actors have entered a new operational phase by deploying novel, autonomous malware, including PROMPTFLUX and PROMPTSTEAL, that leverage Large Language Models mid-execution to dynamically alter ...
▶️ Listen Now