# Contractors Strengthen AI Agent Zero Trust

A Security Info Watch executive Q&A describes controls organizations should establish before deploying autonomous AI agents, including temporary task-specific privileges, privileged access management, segmented security enclaves, immutable audit logs, data-provenance reviews, and testing of system architecture and incident response. For government contractors, these needs point to potential work in Zero Trust modernization, AI-agent governance, security assessments, and tabletop exercises; the signal identifies no specific solicitation, contract, funding, or procurement contact.

- Contractors providing cybersecurity and IT services can align offerings with the controls described, particularly privileged access management, segmentation, audit logging, and agent-focused security assessments.
- Procurement teams evaluating autonomous AI deployments can use these controls as areas to address in security planning and assessment requirements.
- The signal names John Sahlin of General Dynamics Information Technology but does not identify a government buyer, contract vehicle, or funded opportunity.

**Jurisdictions:** federal
**Industries:** Information Technology
**Topics:** Cybersecurity, Artificial Intelligence
**Published:** November 10, 2026

### Vendors
- General Dynamics Information Technology (GDIT) ()

### Sources
- [Executive Q&A: AI Agents Are the New Digital Insiders. Why Zero Trust Must Evolve | Security Info Watch](https://www.securityinfowatch.com/ai/article/55417821/executive-qa-ai-agents-are-the-new-digital-insiders-why-zero-trust-must-evolve) - Security Info Watch