In this episode of Philosophy, Programs, and Prompts, Casey Hart ( @OntologyExplained ) and Carl Brown ( @InternetOfBugs ) break down the real technical story behind OpenAI's bots infiltrating Hugging Face's infrastructure and the subsequent Black Hat conference reveals.Carl explains why claims about LLMs "scheming," creating secret message boards, or building religions are human projections rather than true machine consciousness. We also dive into the myth that you need an LLM to defend against other LLMs, how "honey tokens" work, and why traditional security tools still reign supreme.// Our past episode on the OpenAI HuggingFace Incident:https://www.youtube.com/watch?v=jIlm8O78M2c Subscribe for more grounded, hype-free takes from two real humans who know the tech!More info about Casey:https://CaseyHart.com/More info about Carl:https://InternetOfBugs.com/Show Notes & Topics Discussed:* The OpenAI / Hugging Face security incident and Black Hat presentation* Faithfulness of LLM "Chain of Thought" reasoning logs* The message board myth vs. token caching behavior* Benchmarks like ExploitGym and the Texas Sharpshooter Fallacy* The reality of AI security: Snort, Tripwire, and why LLM firewalls fail// Sources and references:// New Black Hat Hugging Face Break Down Video https://www.youtube.com/watch?v=87DyyMV0kCY// Chain of Thought isn't "Faithful" to what the LLMs actually dohttps://www.anthropic.com/research/reasoning-models-dont-say-thinkhttps://www.anthropic.com/research/measuring-faithfulness-in-chain-of-thought-reasoning//Carl's video on how people said Agents on MoltBook "Created a religion" in the same way they said the agents in this incident "Created a Message Board"https://www.youtube.com/watch?v=GYfgjYVEYQ0// RadioLab Episode that Casey mentioned about how people make "separated at birth" stories more compelling by omitting things:https://radiolab.org/podcast/born-way/transcript// Microsoft's record breaking recent patch release:https://thehackernews.com/2026/07/microsoft-patches-record-622-flaws.htmlTimestamps / Chapter Markers:00:00 - The Fallacy of AI Defending Against AI00:30 - Welcome to Philosophy, Programs, and Prompts01:05 - Re-examining the Hugging Face Security Incident04:26 - Chain of Thought Logs: Are Thought Monologues Faithful?07:30 - Context Windows and Self-Fulfilling Loops10:00 - "Scheming" vs. Next-Token Prediction11:00 - The "AI Message Board" Myth Explained18:05 - Moltbook, Agent Religions, and Media Projection23:00 - Texas Sharpshooter Fallacy in AI Reporting25:35 - Are LLMs Actually Learning?28:40 - What Are Honey Tokens?31:45 - The Myth of the Anomaly Detection LLM34:50 - ExploitGym and Training Defenses vs. Attacks36:50 - Why Traditional Security (Tripwire/Snort) Beats LLMs42:05 - Why Haven't Rogue Frontier Models Broken the Internet?45:34 - Outro and Upcoming Hank Green Episode#ArtificialIntelligence #Cybersecurity #OpenAI #HuggingFace #TechPodcast #LLM #Infosec
Podden och tillhörande omslagsbild på den här sidan tillhör
Casey Hart, Carl Brown. Innehållet i podden är skapat av Casey Hart, Carl Brown och inte av,
eller tillsammans med, Poddtoppen.