Steve Gibson
Security Now (Audio)
Tokens in the Stream - Why LLMs are inherently insecure and prompt injection will persist
- Published
- August 26, 2026
- Duration
- 2h 48m
- Summary source
- description
- Last updated
- Aug 26, 2026
Discusses anthropic, agents, ai-regulation.
Summary
Turns out, every chatbot conversation runs on a messy hack at the heart of language models, making prompt injection an unsolved—and possibly unsolvable—security threat. Steve and Leo unravel the research that explains why "roles" in AI aren't what you think they are. Understanding the controversy surrounding "AI Model Distillation" Anthropic moves to make…
Intelligent Report
Sign in to read teasers, or upgrade to Research Pro to commission intelligent report for this episode. Learn more →
Show notes
Turns out, every chatbot conversation runs on a messy hack at the heart of language models, making prompt injection an unsolved—and possibly unsolvable—security threat. Steve and Leo unravel the research that explains why "roles" in AI aren't what you think they are. Understanding the controversy surrounding "AI Model Distillation" Anthropic moves to make their most powerful Mythos 5 model more widely available. Bitwarden's "Secrets Manager" offering prevents agentic and prompt injection abuse.
Themes
- anthropic
- agents
- ai-regulation