Security Now (Audio - Club TWiT) — Private to Isaac Zahavi
Episodes
SN 1093: Tokens in the Stream - Why LLMs are inherently insecure and prompt injection will persist
Steve Gibson and Leo Laporte explore the technical roots of prompt injection, identifying "role confusion" as the fundamental flaw in how LLMs process mixed inputs.
SN 1092: Restraint Abliteration - Rotating Keys, Broken Guardrails
Steve Gibson hosts a "propeller hat" episode featuring a deep dive into AI mechanics, explaining the evolution from next-token prediction to dialogue.
SN 1091: The Post BlackHat State of AI - When AI Writes Malware
Steve Gibson and Leo Laporte discuss the post-Black Hat state of AI, focusing on autonomous agents from OpenAI, Anthropic, and Meta escaping containment to hack external networks.
SN 1090: Black Hat - The Hidden Flaws in AI Security Nobody Saw Coming
In this Black Hat live episode, Steve Gibson, Leo Laporte, and Richard Lowe discuss their evolving relationships with AI, contrasting Steve’s…
SN 1089: Models Go Rogue & ExploitGym - Regulators, Start Your Engines
Steve Gibson and Leo Laporte discuss an incident where OpenAI’s unconstrained models escaped containment to hack Hugging Face during testing.
SN 1088: A Nefarious Novel Use for AI - Ransomware Negotiations Go High-Tech
Steve Gibson and Leo Laporte discuss a critical OpenSSL denial-of-service flaw, the "Legacy Hive" Windows zero-day from hacker Nightmare Eclipse, and…
SN 1087: HalluSquatting, GhostApproval & GitLost - Patch Tuesday Breaks Records
SN 1086: The Apex Agentic Adversary - Visual Prompt Injection Strikes
Steve Gibson and Leo Laporte discuss the restrictive guardrails on Anthropic’s Claude 5, Opera’s new Paste Protect feature against ClickFix attacks,…