IP
Indirect Prompt Injection
Indexing
Sifting through hundreds of thousands of hours of indexed videos
Indirect Prompt Injection
Sifting through hundreds of thousands of hours of indexed videos
Indirect Prompt Injection
Arcmira media summary
Explore podcasts, interviews & explainers on Indirect Prompt Injection — 8 indexed from David Bombal & OWASP Foundation, updated Apr 2026.
A more complex threat where AI is manipulated via external data sources like documents or tools.
Discussion of how instructions can be injected via MCP servers to exfiltrate data.
The primary technical subject of the talk involving malicious instructions in external data.
The technical method used to hijack AI agents via untrusted data sources.
A security threat where an AI agent processes malicious instructions from third-party data.
Arcmira tracks 8 indexed media appearances or mentions for Indirect Prompt Injection, tied to source videos, channels, and transcript-derived context.
Arcmira uses indexed YouTube videos and transcripts. Representative source evidence on this page includes "Why AI Guardrails Are Dead & The Threat of Indirect Prompt Injection" with transcript-derived context and links when available.
Indirect Prompt Injection is connected to Google, OpenAI, Microsoft in Arcmira's media graph.
8
Mentions
1.6M
Views
The trendline is visible, but the dated evidence behind Indirect Prompt Injection is in the premium layer.
“A more complex threat where AI is manipulated via external data sources like documents or tools.”
“Discussion of how instructions can be injected via MCP servers to exfiltrate data.”
“The primary technical subject of the talk involving malicious instructions in external data.”
“The technical method used to hijack AI agents via untrusted data sources.”
“A security threat where an AI agent processes malicious instructions from third-party data.”