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.
Practice-by-practice evidence
Denominator: 11 records · use cases designated high-impactReported in placeReported in progressNot applicableNo answer reported
Pre-deployment testing
Pre-deployment testing
Reported in place: 1 (9.1%)Pre-deployment testing
No answer reported: 10 (90.9%)
11Reported in place: 1 (9.1%)Pre-deployment testing
No answer reported: 10 (90.9%)
AI impact assessment11
Independent review11
Ongoing monitoring11
Operator training11
User and public consultation
User and public consultation
Reported in place: 1 (9.1%)User and public consultation
No answer reported: 10 (90.9%)
11Reported in place: 1 (9.1%)User and public consultation
No answer reported: 10 (90.9%)
| Practice | In scope | In place | In progress | Not in place | N/A | Precluded or waived | Other | No answer | Answered | In place of applicable |
|---|---|---|---|---|---|---|---|---|---|---|
| Pre-deployment testing | 11 | 1 | 0 | 0 | 0 | 0 | 0 | 10 | 9% | 9% of 11 |
| AI impact assessment | 11 | 1 | 0 | 0 | 0 | 0 | 0 | 10 | 9% | 9% of 11 |
| Independent review | 11 | 1 | 0 | 0 | 0 | 0 | 0 | 10 | 9% | 9% of 11 |
| Ongoing monitoring | 11 | 0 | 1 | 0 | 0 | 0 | 0 | 10 | 9% | 0% of 11 |
| Operator training | 11 | 1 | 0 | 0 | 0 | 0 | 0 | 10 | 9% | 9% of 11 |
| Fail-safe | 11 | 1 | 0 | 0 | 0 | 0 | 0 | 10 | 9% | 9% of 11 |
| Appeal process | 11 | 0 | 0 | 0 | 1 | 0 | 0 | 10 | 9% | 0% of 10 |
| User and public consultation | 11 | 1 | 0 | 0 | 0 | 0 | 0 | 10 | 9% | 9% of 11 |
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.
| Testing | Impact assess. | Indep. review | Monitoring | Training | Fail-safe | Appeal | Consultation | n | |
|---|---|---|---|---|---|---|---|---|---|
| DOJ | – | – | – | – | – | – | – | – | 4 |
| VA | – | – | – | – | – | – | – | – | 3 |
| DOE | – | – | – | – | – | – | – | – | 3 |
| SSA | 100% | 100% | 100% | 0% | 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
11 records with at least one practice unanswered or in progressOpen in explorer| Use case | Stage | Test | AIA | Review | Monitor | Train | Fail-safe | Appeal | Consult |
|---|---|---|---|---|---|---|---|---|---|
| Custom GenAI for eVinci Microreactor Engineering (MauroGPT) DOE · DOE-534 | Pilot | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported |
| Custom GenAI for Advanced Reactor Development (LotusAI) DOE · DOE-591 | Pilot | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported |
| Custom GenAI for Advanced Reactor Development (RickoverAI) DOE · DOE-609 | Pilot | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported |
| Generative AI R&D Sandbox DOJ · DOJ-0090 | Pilot | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported |
| Voicemail Transcription, Translation and Summarization DOJ · DOJ-0243 | Pilot | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported |
| Conduit AI DOJ · DOJ-0310 | Pilot | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported |
| Cocounsel AI DOJ · DOJ-0313 | Pilot | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported |
| Policy Assistance Tool (PAT) SSA · R3116 | Pilot | Reported in place | Reported in place | Reported in place | Reported in progress | Reported in place | Reported in place | Not applicable | Reported in place |
| Abridge Ambient Scribe VA · VA-24-708 | Pilot | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported |
| Knowtex Ambient Scribe VA · VA-25-6426 | Pilot | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported |
| Patient Care Services Integration Platform (PCSIP) VA · VA-24-421 | Pilot | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported | No answer reported |
Unanswered analytical questions
Generated from the data; for follow-up with agencies- DOJ designated 4 use cases high-impact but answered none of the minimum-practice questions for any of them. Were the practices assessed, and where is that recorded?
- VA designated 3 use cases high-impact but answered none of the minimum-practice questions for any of them. Were the practices assessed, and where is that recorded?
- DOE designated 3 use cases high-impact but answered none of the minimum-practice questions for any of them. Were the practices assessed, and where is that recorded?
- 1 high-impact use cases mark an appeal process as not applicable. On what basis was each individual outcome judged not contestable?
- 4 high-impact use cases report using PII but give no link to a Privacy Impact Assessment. Is a PIA published elsewhere?
- 6 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.