We are handing non-deterministic AI agents static API keys, then acting surprised when production databases get wiped.
In this episode of The Secure Disclosure, host Mackenzie Jackson sits down with Eyal Ben Ezra, CEO of Willow, to explore how modern enterprises can safely control autonomous AI agents. They discuss moving beyond traditional permission scopes toward capability-based dynamic guardrails, securing tool calls via MCP gateways, and why Eyal believes prompt injection is fundamentally unsolvable with current technology.
From rogue agents purchasing cloud domains to accidental database catastrophes, learn how to protect critical infrastructure while adopting enterprise AI.
Podden och tillhörande omslagsbild på den här sidan tillhör
Mackenzie Jackson. Innehållet i podden är skapat av Mackenzie Jackson och inte av,
eller tillsammans med, Poddtoppen.
