System prompt
Type: Concept
As of November 2026, System prompt has been mentioned in one podcast episode on The Ezra Klein Show (Oct 2025).
Prompt injection is a cybersecurity exploit and an attack vector in which innocuous-looking inputs are designed to cause unintended behavior in machine learning models, particularly large language models (LLMs). The attack takes advantage of the model's inability to distinguish between developer-defined prompts and user inputs to bypass safeguards and influence model behaviour. While LLMs are designed to follow trusted instructions, they can be manipulated into carrying out unintended responses
| Mentioned in | 1 episode on 1 podcast (as of November 2026) |
|---|---|
| First heard | Oct 2025 on The Ezra Klein Show |
| Field | Artificial Intelligence, Natural Language Processing |
Who talks about System prompt on podcasts?
System prompt came up; the most recent mention was Oct 2025 on The Ezra Klein Show.
What is System prompt mentioned alongside?
System prompt is most often mentioned alongside The New York Times Magazine, Reinforcement learning (RL) and Silicon Valley.
Mentioned together: The New York Times Magazine, Reinforcement learning (RL), Silicon Valley, Ezra Klein, ChatGPT, DeepMind, Claude, AI alignment, Gemini, AlphaGo