Agent trust system boosts security for AI teamwork across companies

Agent Trust Infrastructure (ATI): A Federated Identity, Trust, Attestation, and Accountability Architecture for Cross-Organization Autonomous AI Agents

Artificial Intelligence

Summary

AI agents often need to work together even when run by different companies, but now there isn’t a good way to check if each agent is who it claims to be or if it’s following rules. The authors created a new system called Agent Trust Infrastructure (ATI) that uses certificates and security checks to verify agents reliably without requiring all companies to share one central authority. ATI helps stop bad actions by checking agents’ identities and their tools before they act, and it adds very little delay. The system works with many existing AI setups and is available publicly for others to use.

What this means in practice

  • For enterprise security teams: Protect cross-company AI collaborations by verifying agent identities, tool integrity, and enforcing authorization policies using ATI.
  • For ai platform developers: Integrate ATI externally to existing AI agent frameworks for enhanced trust and security without changing core AI models.

Authors

Viet K. Nguyen, Mohammad I. Husain

Abstract

AI agents increasingly collaborate across organizational boundaries, invoking tools and interacting with agents operated by independent entities. Although existing agent frameworks support planning and coordination, they lack a common mechanism for verifying cross-organizational agent identity, execution integrity, and authorization. We propose the Agent Trust Infrastructure (ATI), a framework-independent security architecture built on Public Key Infrastructure (PKI), Zero Trust Architecture (ZTA), and remote attestation. ATI establishes federated trust through peer organizational certificate authorities (AgentCAs), eliminating the need for a shared root authority. Before executing requests, ATI validates agent certificates, verifies approved model, tool, and system-prompt configurations, and enforces authorization through three policy checks. We implement ATI as an external security enforcement layer that integrates with existing agent frameworks without modifying their underlying models. We evaluate ATI in a simulated federation of eight organizations using six agent frameworks, each deploying a five-agent team. Security experiments involve two organizations and assess 14 attack scenarios and six benign tasks across five agent models, using five attack variations and three repetitions per variation. A panel of judge models evaluates the experimental outcomes. With all enforcement mechanisms enabled, attack containment increases from 43% to 84% (40% to 80% for the 11 attack scenarios evaluated consistently across all configurations), while enforcement introduces approximately 174 microseconds of overhead per request. ATI is released as an open-source, framework-independent package that uses IETF WIMSE workload token formats for identity and request verification and Entity Attestation Tokens for model, tool, and system-prompt integrity verification.