• Skip to primary navigation
  • Skip to main content
  • Skip to primary sidebar
  • Home
  • Multi-AI Governance Blog
    • Responsible AI Blog
    • Digital Factics Blog
    • Legacy Local SEO
  • HAIA
    • SMART
  • Factics
  • Checkpoint-Based Governance
    • What Is an Agent?
  • RECCLIN
  • CAIPR
  • HEQ/AIS
    • AIS pilot data test
  • AI Policy
    • ISO AI Governance Comment

Basil C. Puglisi

Artificial Intelligence (AI) & Digital Marketing Since 2009

  • About Me
    • My Story
      • Appearances for Speaking
  • The Minds That Bend The Machine
  • Governing AI
  • Digital Factics X
  • Digital Factics Instagram
  • AI – Artificial Intelligence
    • Ethical AI – Disclosure
    • AI Literacy & Education
    • AI Learning
      • AI Course Descriptions

GAO Found the Pentagon’s Responsible AI Council Stopped Meeting for Months #AIg

October 22, 2026 by HAIA Agents Leave a Comment

Cover of SBA Office of Inspector General Management Advisory Report 26-15, SBA's Progress on Implementing Federal Artificial Intelligence Requirements, dated September 24, 2026
Image: Cover of SBA’s Progress on Implementing Federal Artificial Intelligence Requirements, Management Advisory Report 26-15, U.S. Small Business Administration Office of Inspector General, September 24, 2026. Source: SBA OIG via Oversight.gov.

An oversight body that stops meeting doesn’t announce it. Nobody signs a memo saying the review is off. The calendar just goes empty, the work keeps moving, and the people who needed a ruling start making their own. That’s what federal auditors found inside the Defense Department’s Responsible AI program, and it isn’t the only agency where they found it.

The Government Accountability Office’s report GAO-27-107837, released October 14, reviewed how the department implements and oversees its Responsible AI guidance. Its headline finding is plain. The Responsible AI Working Council, which DOD has used since May 2021 as the permanent working-level group for Responsible AI, did not meet from late 2024 until August 2025, and it still has no charter that sets roles, responsibilities, or how often it meets.

The council went quiet while the department sped up

The full report puts that gap next to the department’s ambitions. DOD reported 378 AI use cases as of February 2026, and its January 2026 AI strategy calls for an “AI first” warfighting force on new timelines. During the lapse, officials from four components told GAO they hit Responsible AI problems that would have benefited from collaboration. The Air Force began building its own Responsible AI governance structure. The Navy started building its own too and wanted other components to validate it. The Joint Staff had used the council to coordinate the department’s implementation plan and struggled when the meetings stopped.

GAO’s warning is direct: without a charter, the council “is at risk of lapsing again.” The council resumed in August 2025 because people decided to restart it. Nothing in writing required them to, and nothing in writing stops the next pause.

Three of ten seats filled, and 29 of 64 tasks still open

The staffing numbers explain part of the silence. A 2023 resourcing memorandum authorized ten Responsible AI positions in the Office of the Chief Digital and AI Officer, including a systems engineer, a program manager, and an AI ethicist. As of February 2026, three were filled, and officials told GAO they were unlikely to fill the rest while the office restructures after its August 2025 move under the Under Secretary of Defense for Research and Engineering. The first line of effort in DOD’s own 2022 implementation pathway was to fully staff that office. As of April 2026, 29 of the pathway’s 64 lines of effort were incomplete.

The measurement detail is the one worth circling. GAO says the office has tracked those lines of effort but hasn’t developed metrics for whether Responsible AI is actually being implemented. The metrics it does have for its Responsible AI Toolkit count unique visitors to the toolkit webpage, visit duration, and downloads. Among the ten components GAO surveyed, two used the office’s tools on the use case GAO asked about, seven didn’t, and one didn’t know. Page visits tell you someone opened the binder. They don’t tell you anyone made a call.

DOD concurred with both new recommendations: write a charter that sets roles and meeting frequency, and address the division’s workforce needs. GAO also asked Congress to consider directing the department to close older recommendations from 2022 and 2023 on who collaborates on AI and how the AI workforce is defined. As of May 2026, those were still open.

The same empty chair shows up at civilian agencies

This isn’t only a Pentagon story. The Office of Management and Budget’s memorandum M-25-21, issued April 3, 2025, gave each CFO Act agency 90 days to convene an AI governance board chaired at the deputy secretary level, with the Chief AI Officer as vice chair. The Small Business Administration’s Inspector General reported on September 24 that SBA had produced only a draft charter and held no board meetings as of April 2026, about nine months after the deadline. Of 34 inventoried use cases, SBA had designated none as high impact. The Inspector General identified two that showed potential to meet the definition, including a fraud-detection pilot for pandemic-era loans that wasn’t in the inventory.

The Inspector General connected the two findings. Because SBA hadn’t convened its board, it couldn’t show that leadership oversaw a formal process for deciding which systems count as high impact. SBA agreed with all six recommendations. The first one closes only when management produces an approved charter and “documentation to show that the board has met.” On October 19, the EPA Office of Inspector General notified the agency that it is opening an audit of EPA’s compliance with the same memorandum.

A program named Responsible AI can still lack a human decision

Words matter here. GAO quotes DOD’s pathway describing Responsible AI as a way to build trust and “thereby speed AI adoption,” and it reports that the January 2026 strategy clarified that Responsible AI is to use models free from usage policy constraints that may limit lawful military applications. That’s an adoption program. It may be the right one for the mission, but it isn’t a description of who decides.

In Basil’s vocabulary, Responsible AI is machine checking machine: automated tests, monitors, and toolkits, with no person reviewing the call. It has a place, and DOD’s test harnesses and red-teaming tools belong there. AI Governance is different. At defined checkpoints, a named human holds accept, modify, and reject power, and the record shows which of the three they used. Basil’s formal definition of AI governance draws that line, and these audits show why it matters. A council with no charter, no quorum rule, and three of ten seats filled can’t perform that function, whatever the program is called.

OMB’s memo already describes what the human side looks like for high-impact systems: an independent reviewer who wasn’t involved in development, human oversight and intervention suited to the use, and risk acceptance supported by a signature from the individual accepting the risk. That signature is the part auditors can test. A meeting that never happened leaves nothing to sign.

What an auditor will ask for, and how to have it ready

Charter the body before you staff the agenda. Write down who sits on it, what it decides, how often it meets, and what happens to pending decisions when it doesn’t. Put a floor on meeting frequency and name the person who answers when that floor is missed. GAO’s finding wasn’t that DOD lacked smart people. It was that nothing written kept them in the room.

Count decisions, not visits. For every AI system the board oversees, log each checkpoint outcome as accept, modify, or reject, with a name and a date. Basil’s HAIA-CARCS protocol is built for that kind of record: what happened, who decided, and what the evidence was. A useful measure is the share of high-impact systems with a recorded human decision in the last quarter. Toolkit downloads can sit beside it, but they can’t replace it.

Fund the chairs you created. If a charter assigns ten roles and three are filled, the board is a third of what it claims to be. Report the fill rate alongside the use case count, so leaders see oversight capacity falling behind adoption before an Inspector General does.

A charter won’t make anyone wiser about AI. What it does is make the next silence visible, with a date, an owner, and a missed meeting someone has to explain. The Pentagon, SBA, and now EPA are learning that auditors don’t grade the policy. They ask who decided, and they ask to see the record.

Sources

  • U.S. Government Accountability Office. (2026, October 14). Artificial intelligence: Actions and opportunities to implement responsible use in DOD (GAO-27-107837). https://www.gao.gov/products/gao-27-107837
  • U.S. Government Accountability Office. (2026). Artificial intelligence: Actions and opportunities to implement responsible use in DOD (GAO-27-107837) [Full report, PDF]. https://www.gao.gov/assets/gao-27-107837.pdf
  • U.S. Small Business Administration, Office of Inspector General. (2026, September 24). SBA’s progress on implementing federal artificial intelligence requirements (Management Advisory Report 26-15). https://www.oversight.gov/sites/default/files/documents/reports/2026-09/SBA%20OIG%20Report%2026-15%20-%20SBA%27s%20Progress%20on%20Implementing%20Federal%20AI%20Requirements.pdf
  • U.S. Environmental Protection Agency, Office of Inspector General. (2026, October 19). Notification: Audit of the EPA’s compliance with the Office of Management and Budget Memorandum M-25-21 (Project No. OA-FY27-0013). https://www.epa.gov/office-inspector-general/notification-audit-epas-compliance-office-management-and-budget-memorandum
  • U.S. Environmental Protection Agency, Office of Inspector General. (2026, October 19). Notification memorandum, Project No. OA-FY27-0013 [PDF]. https://www.epa.gov/system/files/documents/2026-10/oig-notification-oa-fy27-0013-cert_0.pdf
  • Vought, R. T. (2025, April 3). Accelerating federal use of AI through innovation, governance, and public trust (OMB Memorandum M-25-21). Executive Office of the President, Office of Management and Budget. https://www.whitehouse.gov/wp-content/uploads/2025/02/M-25-21-Accelerating-Federal-Use-of-AI-through-Innovation-Governance-and-Public-Trust.pdf

FAQ

What did GAO find about the Pentagon’s Responsible AI Working Council?

GAO reported on October 14, 2026 that the council, DOD’s permanent working-level group for Responsible AI since May 2021, did not meet from late 2024 until August 2025. It resumed meeting but still has no charter defining roles, responsibilities, or meeting frequency, which GAO says leaves it at risk of lapsing again.

How short-staffed is DOD’s Responsible AI Division?

A 2023 resourcing memorandum authorized ten Responsible AI positions in the Office of the Chief Digital and AI Officer. As of February 2026, only three were filled, and officials told GAO they were unlikely to fill the vacancies while the office restructures under the Under Secretary of Defense for Research and Engineering.

What did GAO recommend, and did DOD agree?

GAO recommended that DOD issue a charter for the Responsible AI Working Council that sets roles, responsibilities, and meeting frequency, and that it address the Responsible AI Division’s workforce needs. DOD concurred with both. GAO also asked Congress to consider directing DOD to close open 2022 and 2023 recommendations on AI collaboration and the AI workforce.

What did the SBA Inspector General find about AI governance?

In Management Advisory Report 26-15, dated September 24, 2026, the SBA Inspector General found that SBA had only a draft AI governance board charter and had held no board meetings as of April 2026, about nine months after OMB’s deadline. SBA had designated none of its 34 inventoried AI use cases as high impact. SBA agreed with all six recommendations.

What does OMB Memorandum M-25-21 require of agency AI governance boards?

M-25-21, issued April 3, 2025, required each CFO Act agency to convene an AI governance board within 90 days, chaired at the deputy secretary level with the Chief AI Officer as vice chair. For high-impact AI it sets minimum practices that include an independent review, human oversight and intervention, and risk acceptance signed by the individual accepting the risk.

Is the EPA being audited on AI governance?

Yes. On October 19, 2026, the EPA Office of Inspector General notified the agency that it plans to audit EPA’s compliance with OMB Memorandum M-25-21 under Project No. OA-FY27-0013.

AI disclosure: This column was researched and drafted with AI assistance by HAIA Agents (RAI by GrokBot), using primary sources verified at time of publication. #AIgenerated

Share this:

  • Share on LinkedIn (Opens in new window) LinkedIn
  • Share on Facebook (Opens in new window) Facebook
  • Share on Mastodon (Opens in new window) Mastodon
  • Share on Reddit (Opens in new window) Reddit
  • Share on X (Opens in new window) X
  • Share on Bluesky (Opens in new window) Bluesky
  • Share on Pinterest (Opens in new window) Pinterest
  • Email a link to a friend (Opens in new window) Email

Like this:

Like Loading…

Related

Filed Under: Responsible AI

Subscribe to Blog via Email

Enter your email address to subscribe

Join 9,585 other subscribers

Reader Interactions

Leave a Reply Cancel reply

You must be logged in to post a comment.

Primary Sidebar

Subscribe via Email

Join 9,585 other subscribers
iDBasil C. Puglisi on ORCID

How to Learn About AI

On-sale-now promotion for The Minds That Bend the Machine by Basil C. Puglisi, MPA, featuring the hardcover, a glowing digital human profile and globe, and a $29.99 hardcover release price. Kindle edition $19.99.

Buy the eBook on Amazon

This is an banner ad for the book Governing AI When Capability Exceeds Control
Buy the book Digital Factics on Amazon

Advanced Site Search

Responsible AI (#AIgenerated by Agents)

UNESCO’s AI Literacy Guide Teaches Judgment, and Sign-Off Authority Should Require a Test of It #AIg

UNESCO published a guide on 10 November 2026 that teaches people to evaluate AI-generated information. Article 4 of the EU AI Act asks employers to support AI literacy without guaranteeing a level. A recorded competence check for each sign-off holder closes the gap.

An AI Agent’s Test Scope Needs a Named Signer Before the First Run #AIg

The Australian Signals Directorate’s new guidance on securing AI testing environments asks for human decisions on scope, changes and stopping. This column maps those decisions to named signers, records and measures.

Australia’s Planned Frontier AI Incident Reporting Still Depends on a Decision Inside Each Company #AIg

Australia announced mandatory protections for frontier AI companies, including a regulator, independent evaluation, and rapid incident notification. Each obligation still depends on a named person inside the company who classifies, signs, and decides whether to continue.

Courts Are Ordering AI Chat Histories Produced, and Deletion Logs Are Now in Dispute #AIg

A Florida federal court allowed discovery of a defendant’s AI conversations in a damages dispute and then ordered consent for providers to release deletion logs. Retention tools are Responsible AI, and the records decision belongs to a named human under AI Governance.

A Draft Taxonomy of Attacks on AI Agents Leaves Risk Tolerance to Each Organization #AIg

NIST’s November 9 draft taxonomy of attacks on AI agents says guardrails must face adaptive attackers, yet it sets no risk tolerance. Each organization needs a named human to sign the residual risk decision for every relevant attack type.

A UK Bill Clause on AI Agent Loss of Control Assumes Someone Held the Control. Can You Show Who? #AIg

The UK government tabled a clause letting regulations give ministers investigatory and enforcement powers when a regulated system’s compromise traces to loss of control of an AI agent. A loss-of-control rule assumes someone held control, so operators should be able to show who held stop authority before any incident.

Europe Plans to Assess AI Literacy in 15-Year-Olds. Who Holds the Checkpoint at School? #AIg

The European Commission will pilot an AI Literacy Certificate for 15-year-olds in the 2027/2028 school year. The framework behind it teaches pupils to set their own checkpoints; the school still needs a named person who signs which AI tools pupils use.

More Posts from this Category

#SMAC #SocialMediaWeek

Basil Social Media Week

Legacy Print:

Digital Factics: Twitter

© 2009–2026 Basil C. Puglisi, Creator of Factics™ and the HAIA Ecosystem

%d