---
title: "AI agent credentials live in the same box as untrusted code. Two new architectures show where the blast radius actually stops."
date: 2026-04-10
source: http://www.bing.com/news/apiclick.aspx?ref=FexRss&aid=&tid=69d98769a6a74b0ba872100c3d2b435d&url=https%3a%2f%2fventurebeat.com%2fsecurity%2fai-agent-zero-trust-architecture-audit-credential-isolation-anthropic-nvidia-nemoclaw&c=13505049294162098939&mkt=en-us
description: "Anthropic and Nvidia have shipped the first zero-trust AI agent architectures — and they solve the credential exposure problem in opposite ways, with different blast radius implications for enterprise ..."
---

# AI agent credentials live in the same box as untrusted code. Two new architectures show where the blast radius actually stops.

Anthropic and Nvidia have shipped the first zero-trust AI agent architectures — and they solve the credential exposure problem in opposite ways, with different blast radius implications for enterprise ...

*Published: 2026-04-10*