
The Lethal Trifecta of AI Security with Dan Barahona
Om avsnittet
In this episode of The Secure Disclosure, we sit down with 25-year cybersecurity veteran Dan Barahona, founder of AI Security University, to navigate the wild frontier of artificial intelligence security. Dan breaks down the "lethal trifecta" private data, untrusted content, and external communication and warns why giving autonomous agents unchecked permissions can trigger catastrophic internal breaches. We explore the emerging risks of "vibe coding," prompt injection exploits, and how the Model Context Protocol (MCP) equips large language models with real-world execution capabilities. Dan also highlights why organization-wide workforce AI literacy is quickly becoming critical under regulations like the EU AI Act and ISO 42001.
Fler avsnitt
Visa alla avsnitt av The Secure DisclosureThe Secure Disclosure med Mackenzie Jackson finns tillgänglig på flera plattformar. Informationen på denna sida kommer från offentliga podd-flöden.