Photo: U.S. Secretary of Defense, CC BY 2.0 (https://creativecommons.org/licenses/by/2.0), via Wikimedia Commons (source)
The Government Accountability Office reported Wednesday, Oct. 14, that the Defense Department has begun putting Responsible Artificial Intelligence rules into practice on surveyed projects, while collaboration gaps and thin staffing still slow the work, according to GAO-27-107837.
Department guidance created AI Ethical Principles and six Responsible AI tenets, GAO said. Of 10 components surveyed, most told auditors they had started oversight structures and begun applying those tenets on the selected projects reviewed.
Working council lapse
Inside CDAO, the Responsible AI Division set up a working-level Responsible AI Working Council. CDAO officials told GAO that group did not convene from late 2024 through August 2025. Four component offices said they hit implementation problems in that window that collaboration would have eased. Six of the 10 surveyed components said sharing playbooks and tools would help their Responsible AI work.
Meetings restarted in August 2025, officials said, yet GAO found the council still lacks a charter spelling out roles, membership and how often it is to meet — leaving another long pause possible. Without that forum, GAO said, uneven readings of guidance could go unresolved.
Staffing shortfalls
CDAO officials listed further steps that would advance Responsible AI but said vacancies blocked them. As of February 2026, three of 10 authorized seats in the Responsible AI Division were filled, those officials told GAO. The office argues that Responsible AI lets the department field AI faster; filling the division would support that aim, the report said.
Earlier GAO work urged DOD to map who owns AI collaboration across the department and to assign a lead and timeline for defining the AI workforce. Those items remain open. Completing them, GAO said, would better align the department with its AI goals; the office asked Congress to consider directing that follow-through.
Recommendations
Beyond the matter for Congress, GAO told the secretary of defense to have the chief digital and artificial intelligence officer issue a council charter and fix the Responsible AI Division's staffing. DOD concurred with both new recommendations, the report said.
Senate Report 118-188, tied to a Fiscal Year 2025 defense authorization bill, prompted the review. Auditors surveyed 10 components on 10 use cases, examined Responsible AI documents and interviewed officials.
What's next
The two new recommendations and the congressional matter are open. GAO said it will refresh status once it confirms the department's actions. The published report runs 24 pages.
Sources
- U.S. Government Accountability Office: Artificial Intelligence — Actions and Opportunities to Implement Responsible Use in DOD (GAO-27-107837) (2026-10-14)
Motion Media News corrects errors promptly. To report one, email hello@motionnews.studio.
