AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
18min chapter
Coercing LLMs to Do and Reveal (Almost) Anything with Jonas Geiping - #678
The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)
Security Implications of Large Language Models in Agentic Systems
The chapter explores the security risks of using large language models (LLMs) in agentic systems, showcasing their vulnerability to external manipulation for both intended and unintended actions. It discusses attacks on open models, transferability to Blackbots models, and the role of open weight models in security research. The conversation also delves into generative AI games focused on exfiltrating data from LLMs, optimizing attacks using invisible strings, and exploiting system tokens to manipulate model responses.
00:00
Transcript
Episode notes
Get the Snipd
podcast app
Unlock the knowledge in podcasts with the podcast player of the future.
AI-powered
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode
Save any
moment
Hear something you like? Tap your headphones to save it with AI-generated key takeaways
Share
& Export
Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more
AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode