Governance and accountability

Governance evidence

What agencies reported about the OMB M-25-21 minimum practices for high-impact AI. This is a disclosure analysis, not an audit, a compliance finding or an official assessment. An unanswered question is missing evidence, not evidence that a safeguard is absent.

Governance report

Practice-by-practice evidence

Denominator: 14 records · use cases that answered at least one practice question

Reported in placeReported in progressNot applicableWaived by CAIOOther or ambiguous answerNo answer reported

Pre-deployment testing

Pre-deployment testing
Reported in place: 6 (42.9%)Pre-deployment testing
Reported in progress: 5 (35.7%)Pre-deployment testing
No answer reported: 3 (21.4%)

AI impact assessment

AI impact assessment
Reported in place: 5 (35.7%)AI impact assessment
Reported in progress: 5 (35.7%)AI impact assessment
No answer reported: 4 (28.6%)

Independent review

Independent review
Reported in place: 8 (57.1%)Independent review
Reported in progress: 4 (28.6%)Independent review
Waived by CAIO: 1 (7.1%)Independent review
No answer reported: 1 (7.1%)

Ongoing monitoring

Ongoing monitoring
Reported in place: 9 (64.3%)Ongoing monitoring
Reported in progress: 5 (35.7%)

Operator training

Operator training
Reported in place: 8 (57.1%)Operator training
Reported in progress: 6 (42.9%)

Fail-safe

Fail-safe
Reported in place: 7 (50.0%)Fail-safe
Reported in progress: 2 (14.3%)Fail-safe
Not applicable: 5 (35.7%)

Appeal process

Appeal process
Reported in place: 2 (14.3%)Appeal process
Reported in progress: 5 (35.7%)Appeal process
Not applicable: 6 (42.9%)Appeal process
No answer reported: 1 (7.1%)

User and public consultation

User and public consultation
Reported in place: 5 (35.7%)User and public consultation
Reported in progress: 3 (21.4%)User and public consultation
Other or ambiguous answer: 2 (14.3%)User and public consultation
No answer reported: 4 (28.6%)

Practice	In scope	In place	In progress	N/A	Precluded or waived	Other	No answer	Answered	In place of applicable
Pre-deployment testing	14	6	5	0	0	0	3	79%	43% of 14
AI impact assessment	14	5	5	0	0	0	4	71%	36% of 14
Independent review	14	8	4	0	1	0	1	93%	62% of 13
Ongoing monitoring	14	9	5	0	0	0	0	100%	64% of 14
Operator training	14	8	6	0	0	0	0	100%	57% of 14
Fail-safe	14	7	2	5	0	0	0	100%	78% of 9
Appeal process	14	2	5	6	0	0	1	93%	25% of 8
User and public consultation	14	5	3	0	0	2	4	71%	36% of 14

Answered = records with any answer ÷ records in scope. In place of applicable = reported in place ÷ (in scope minus not applicable, precluded and waived). Unanswered records stay in that denominator, so the rate is a lower bound on reported evidence, not an estimate of real practice.

Agency-by-safeguard heatmap

Cell: share reported in place, among applicable recordsIn place Answered

Agency by governance practice
	Testing	Impact assess.	Indep. review	Monitoring	Training	Fail-safe	Appeal	Consultation	n
DOE	63%	50%	57%	63%	50%	100%	67%	50%	8
DOI	–	–	100%	100%	100%	0%	0%	–	3
DOJ	0%	0%	0%	0%	0%	0%	0%	0%	2
FERC	100%	100%	100%	100%	100%	100%	0%	100%	1

Hatched cells: the agency answered nothing for that practice, so there is no evidence either way. Compare agencies with care: each agency is one reporting unit, and portfolios differ in size and type.

Records requiring further investigation

10 records with at least one practice unanswered or in progressOpen in explorer

Use case	Stage	Test	AIA	Review	Monitor	Train	Fail-safe	Appeal	Consult
AI for Operations Center DOE · DOE-10	Deployed	Reported in place	No answer reported	No answer reported	Reported in place	Reported in place	Reported in place	No answer reported	Reported in place
GitHub Copilot DOE · DOE-586	Deployed	Reported in place	Reported in place	Reported in place	Reported in place	Reported in progress	Reported in place	Not applicable	Reported in place
Scripting DOE · DOE-674	Deployed	Reported in progress	Reported in progress	Reported in progress	Reported in progress	Reported in progress	Reported in place	Not applicable	Reported in place
DocketScope DOI · DOI-0040	Deployed	No answer reported	No answer reported	Reported in place	Reported in place	Reported in place	Not applicable	Reported in progress	No answer reported
Dragon DOJ · DOJ-0057	Deployed	Reported in progress	Reported in progress	Reported in progress	Reported in progress	Reported in progress	Reported in progress	Reported in progress	Reported in progress
Dragon DOJ · DOJ-0196	Deployed	Reported in progress	Reported in progress	Reported in progress	Reported in progress	Reported in progress	Reported in progress	Reported in progress	Reported in progress
Microsoft Copilot (Pilot) DOE · DOE-603	Pilot	Reported in progress	Reported in progress	Reported in place	Reported in progress	Reported in progress	Reported in place	Not applicable	Reported in progress
AI workflow to process voicemail user service requests and translate them into actionable IT service tickets DOE · DOE-729	Pre-deployment	Reported in progress	Reported in progress	Reported in progress	Reported in progress	Reported in progress	Not applicable	Not applicable	No answer reported
LLM-Assisted Volcanic Alert Monitoring DOI · DOI-0109	Pilot	No answer reported	No answer reported	Reported in place	Reported in place	Reported in place	Not applicable	Reported in progress	No answer reported
ATLAS DOI · DOI-0077	Pre-deployment	No answer reported	No answer reported	Reported in place	Reported in place	Reported in place	Not applicable	Reported in progress	No answer reported

Unanswered analytical questions

Generated from the data; for follow-up with agencies

1 governance answers are ambiguous (for example, an answer option from a different question). What did the agency intend?

These are questions the inventory raises but cannot answer. They are not findings of non-compliance.