# Federal Agencies Prioritize AI Data Integrity

Federal agencies including CISA, NSA, FBI, NIST, and DoD are emphasizing the critical need to address data poisoning risks in AI systems, particularly large language models used in national security and critical infrastructure. Procurement policies are urged to incorporate stringent data provenance verification, adversarial training techniques, and supply chain risk management to safeguard AI deployments. This focus reflects a strategic priority to ensure AI system resilience and integrity comparable to software supply-chain security efforts.

- Agencies must integrate data integrity requirements into AI-related procurements to mitigate adversarial manipulation risks.
- Contractors providing AI solutions should demonstrate capabilities in secure data sourcing, adversarial robustness, and supply chain transparency.
- This emphasis signals growing demand for technologies and services that enhance AI system trustworthiness and resilience.
- Procurement professionals should align acquisition strategies with federal guidance to support secure AI adoption in defense and critical infrastructure sectors.

**Jurisdictions:** federal
**Industries:** Defense & Military, Information Technology
**Topics:** Cybersecurity, Artificial Intelligence
**Published:** September 23, 2026

### Government Entities
- Cybersecurity and Infrastructure Security Agency (CISA)
- National Security Agency (NSA)
- Federal Bureau of Investigation (FBI)
- National Institute of Standards and Technology (NIST)
- Department of Defense (DoD)

### Key Quotes
> Data integrity for AI should be given the same strategic priority by the US government as software supply-chain security and critical infrastructure protection.
> — Chuck Brooks, President of Brooks Consulting International

### Sources
- [AI & the Threats & Dangers of Data Poisoning](https://www.govconwire.com/articles/chuck-brooks-govcon-expert-ai-data-poisoning-training-national-security) - GovCon Wire