k
keynews.ai
keynews.ai
AgentsThe story, in brief

How Benchling secured multi-tenant AI agents with Amazon Bedrock AgentCore

Not a pilot. Benchling deployed AI agents running untrusted code across thousands of life-sciences tenants—here's how they locked it down.

Reporting by AWS Machine Learning Blog 1 min brief
Illustration of independent geometric mechanisms passing paper tasks along branching amber tracks.
AI agents and the coordination of work.AI illustration by KeyNews
The KeyNews take

Why it matters

Multi-tenant agent security is becoming a solved problem. Benchling's defense-in-depth architecture (sandboxing + DNS firewall + VPC policies) shows how to run autonomous code generation safely at scale—a blueprint for enterprises deploying agents across customer data.

The key facts

11 to know

  1. Benchling deployed AI agents across thousands of life-sciences tenants

  2. Used Amazon Bedrock AgentCore Code Interpreter in VPC mode for sandboxing

  3. Combined Route 53 Resolver DNS Firewall with VPC endpoint policies to block data exfiltration

  4. Architecture addresses untrusted, agent-generated code execution

  5. Defense-in-depth approach: sandboxing + DNS controls + network policies

  6. Benchling deployed AI agents across thousands of tenants

  7. Used Amazon Bedrock AgentCore Code Interpreter in VPC mode

  8. Implemented defense-in-depth: Route 53 DNS Firewall + VPC endpoint policies

  9. Focused on blocking data exfiltration, including DNS-based attacks

  10. Life sciences / scientific code execution as use case

  11. Multi-tenant isolation and untrusted code execution as core problem

Go to the source

AWS Machine Learning Blogaws.amazon.com

Publisher excerpt: Learn how Benchling built a defense-in-depth security architecture to run untrusted, AI agent-generated scientific code across thousands of life sciences tenants using Amazon Bedrock AgentCore Code Interpreter in VPC mode, combined with Amazon Route 53 Resolver DNS Firewall and VPC endpoint…
Read original report
Back to today's editionMore agents news

The wider picture

View all
Illustration of independent geometric mechanisms passing paper tasks along branching amber tracks.
AI illustration by KeyNews
Agents01

On theCUBE Pod: Dreamforce unveils the agentic dream, and Nscale goes public

Salesforce's AIforce represents a major vendor push into agentic workflows at scale. Combined with Nscale's public market debut, this signals enterprise agent adoption is moving from pilot to production—and it's reshaping how vendors compete.

SiliconAngle
Illustration of independent geometric mechanisms passing paper tasks along branching amber tracks.
AI illustration by KeyNews
Agents02

Multi-agent AI systems are taking over supply chain execution

Enterprise logistics is moving from human-approved AI recommendations to fully autonomous agent execution across operational workflows. This represents a real shift in where autonomy is being deployed and how it's changing decision velocity.

AI News
Illustration of independent geometric mechanisms passing paper tasks along branching amber tracks.
AI illustration by KeyNews
Agents03

AI Security Is an Engineering Problem — How to Solve It at Every Layer of the Agent Stack

As agents move into production, security can't be bolted on as an afterthought. This piece articulates a framework for security-by-design across the agent stack—defining requirements, controls, and ownership patterns that practitioners need to adopt now.

NVIDIA Blog