US DoD Chief Digital and AI Officer Craig Martell says the agency wants to use AI safely and urges companies to share insights into how their AI tools are built
Context & Ripple Effects
The request extends the DoD’s earlier ethical principles for AI use from high-level guidance toward practical visibility into the systems it may use. It also follows the Pentagon’s creation of a dedicated digital and AI leadership role with Martell’s appointment as its first chief digital and AI officer.
The significance is not a new AI deployment but a procurement-facing signal: safety is being tied to vendors’ willingness to explain how their tools are made, rather than treated solely as an internal government responsibility.
First-order effects
- AI companies seeking to work with the DoD are being asked to provide more insight into how their AI tools are built.
- DoD AI teams have a clearer stated basis for evaluating safety alongside a supplier’s technical transparency.
Second-order effects
- The request can favor vendors able to document and explain their systems, while raising the bar for less transparent AI offerings in defense-facing engagements.
- If incorporated into evaluations, supplier disclosures could make assurance and auditability more salient competitive features than raw model capability alone.
Third-order effects
- If this approach is repeated across government AI buying, public-sector procurement could become a channel for setting operational transparency norms in dual-use AI.
- The longer-term shift would be from voluntary ethics statements toward evidence-based AI assurance, though this report does not establish a binding disclosure requirement.
The trend: This is one data point in the shift toward government AI adoption that conditions access on demonstrable safety and transparency.