# Defense Organizations Strengthen AI Assurance

Seekr President Rob Clark highlighted security and assurance risks in open-weight AI models used in defense, including data poisoning, backdoors, and compromised data lineage. For procurement teams and defense contractors, these risks point to potential demand for model-provenance verification, predeployment AI testing, and tools that detect spoofing or manipulation in disconnected, high-consequence operating environments.

- Acquisition teams can account for model provenance, security testing, and manipulation detection when evaluating AI systems for defense use.
- Contractors may find opportunities to provide assurance tools that address these risks, particularly for systems operating in disconnected environments.

**Jurisdictions:** federal
**Industries:** Defense & Military, Information Technology
**Topics:** Artificial Intelligence, Cybersecurity
**Published:** October 06, 2026

### Vendors
- Seekr ()

### Sources
- [Seekr’s Rob Clark on building trust in autonomous systems | DefenseScoop](https://defensescoop.com/video/seekrs-rob-clark-on-building-trust-in-autonomous-systems) - DefenseScoop