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: 21 records · all use cases in the current filterReported in place: 17 (81.0%)Pre-deployment testing
No answer reported: 4 (19.0%)
Reported in place: 15 (71.4%)AI impact assessment
Reported in progress: 2 (9.5%)AI impact assessment
No answer reported: 4 (19.0%)
| 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 | 21 | 17 | 0 | 0 | 0 | 0 | 0 | 4 | 81% | 81% of 21 |
| AI impact assessment | 21 | 15 | 2 | 0 | 0 | 0 | 0 | 4 | 81% | 71% of 21 |
| Independent review | 21 | 19 | 2 | 0 | 0 | 0 | 0 | 0 | 100% | 90% of 21 |
| Ongoing monitoring | 21 | 21 | 0 | 0 | 0 | 0 | 0 | 0 | 100% | 100% of 21 |
| Operator training | 21 | 21 | 0 | 0 | 0 | 0 | 0 | 0 | 100% | 100% of 21 |
| Fail-safe | 21 | 13 | 0 | 0 | 8 | 0 | 0 | 0 | 100% | 100% of 13 |
| Appeal process | 21 | 5 | 4 | 0 | 12 | 0 | 0 | 0 | 100% | 56% of 9 |
| User and public consultation | 21 | 11 | 0 | 0 | 0 | 0 | 6 | 4 | 81% | 52% of 21 |
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 | |
|---|---|---|---|---|---|---|---|---|---|
| DHS | 100% | 94% | 94% | 100% | 100% | 100% | 100% | 69% | 16 |
| DOI | – | – | 100% | 100% | 100% | 0% | 0% | – | 4 |
| USDA | 100% | 0% | 0% | 100% | 100% | 100% | 100% | 0% | 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
6 records with at least one practice unanswered or in progressOpen in explorer| Use case | Stage | Test | AIA | Review | Monitor | Train | Fail-safe | Appeal | Consult |
|---|---|---|---|---|---|---|---|---|---|
| Mobile Fortify DHS · DHS-2731 | Deployed | Reported in place | Reported in progress | Reported in progress | Reported in place | Reported in place | Reported in place | Not applicable | Reported in place |
| U.S. Wind Turbine Database DOI · DOI-0080 | Deployed | No answer reported | No answer reported | Reported in place | Reported in place | Reported in place | Not applicable | Reported in progress | No answer reported |
| Video Surveillance System USDA · USDA-025 | Deployed | Reported in place | Reported in progress | Reported in progress | Reported in place | Reported in place | Reported in place | Reported in place | Other or ambiguous answer |
| Flor-AI: Developing a Remotely Sensed Image Classification Method for Inventory and Monitoring of Flora in Digital UAS Imagery DOI · DOI-0079 | Pilot | No answer reported | No answer reported | Reported in place | Reported in place | Reported in place | Not applicable | Reported in progress | No answer reported |
| Automating the Detection and Classification of Wildlife in Aerial Imagery [2024 INV#WO0000000109409] DOI · DOI-0075 | Pilot | No answer reported | No answer reported | Reported in place | Reported in place | Reported in place | Not applicable | Reported in progress | No answer reported |
| Automated photographic identification of Eastern box turtles DOI · DOI-0073 | Pilot | 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- 2 high-impact use cases are reported as deployed while their AI impact assessment is reported in progress. Was deployment authorised before the assessment was complete?
- 12 high-impact use cases mark an appeal process as not applicable. On what basis was each individual outcome judged not contestable?
- 2 high-impact use cases report using PII but give no link to a Privacy Impact Assessment. Is a PIA published elsewhere?
These are questions the inventory raises but cannot answer. They are not findings of non-compliance.