{
  "schema_version": "rangoon.article.v1",
  "id": "https://rangoon.ai/insights/archive-omb-m-25-22-ai-procurement-april-2025/",
  "url": "https://rangoon.ai/insights/archive-omb-m-25-22-ai-procurement-april-2025/",
  "markdown_url": "https://rangoon.ai/insights/archive-omb-m-25-22-ai-procurement-april-2025/index.md",
  "json_url": "https://rangoon.ai/insights/archive-omb-m-25-22-ai-procurement-april-2025/index.json",
  "title": "OMB M-25-22 treated AI procurement as an acquisition evidence problem",
  "summary": "OMB Memorandum M-25-22, dated April 3, 2025, addressed efficient federal AI acquisition and separated purchasing discipline from agency-use governance.",
  "language": "en",
  "author": {
    "name": "Rangoon Editorial",
    "url": "https://rangoon.ai/insights/editorial/"
  },
  "publisher": {
    "name": "Rangoon",
    "maintainer": "Hypler",
    "url": "https://rangoon.ai/"
  },
  "format": "News analysis",
  "topic": {
    "slug": "governance",
    "label": "Governance",
    "url": "https://rangoon.ai/insights/topics/governance/"
  },
  "tags": [
    {
      "slug": "omb",
      "label": "OMB",
      "url": "https://rangoon.ai/insights/tags/omb/"
    },
    {
      "slug": "ai-procurement",
      "label": "AI procurement",
      "url": "https://rangoon.ai/insights/tags/ai-procurement/"
    },
    {
      "slug": "federal-ai",
      "label": "federal AI",
      "url": "https://rangoon.ai/insights/tags/federal-ai/"
    },
    {
      "slug": "government-acquisition",
      "label": "government acquisition",
      "url": "https://rangoon.ai/insights/tags/government-acquisition/"
    }
  ],
  "dates": {
    "published": "2026-10-23",
    "updated": "2026-10-23",
    "event": "2025-04-03"
  },
  "event_source_id": 1,
  "retrospective": true,
  "reading_minutes": 3,
  "takeaways": [
    "M-25-22 is about acquiring AI; governance approval of a use case remains a separate decision.",
    "Procurement records should bind requirements to measurable acceptance evidence.",
    "A model claim, a contract award and an operating service describe different states."
  ],
  "source_note": "Historical reporting: OMB Memorandum M-25-22 is dated April 3, 2025 and provides guidance on efficient acquisition of AI in government. This article treats it as a procurement-policy record, distinct from M-25-21’s agency-use governance guidance.",
  "sources": [
    {
      "id": 1,
      "title": "Office of Management and Budget: M-25-22 Driving Efficient Acquisition of Artificial Intelligence in Government",
      "url": "https://www.whitehouse.gov/wp-content/uploads/2025/02/M-25-22-Driving-Efficient-Acquisition-of-Artificial-Intelligence-in-Government.pdf",
      "published": "2025-04-03"
    },
    {
      "id": 2,
      "title": "GovInfo: M-25-22 Driving Efficient Acquisition of Artificial Intelligence in Government",
      "url": "https://www.govinfo.gov/content/pkg/GOVPUB-PREX2-PURL-gpo244619/pdf/GOVPUB-PREX2-PURL-gpo244619.pdf",
      "published": "2025-04-03"
    }
  ],
  "image": {
    "url": "https://rangoon.ai/assets/editorial/archive-omb-m-25-22-ai-procurement-april-2025-diagram-v1.webp",
    "thumbnail_url": "https://rangoon.ai/assets/editorial/archive-omb-m-25-22-ai-procurement-april-2025-diagram-v1-thumb.webp",
    "alt": "Original policy review diagram: AI acquisition evidence lifecycle. Concepts: Requirements, Offer evidence, Contract terms, Acceptance test, Enabled service, Change notice, Exit record. Editorial checklist, not vendor test results or an official legal determination.",
    "type": "original-diagram",
    "caption": "AI acquisition evidence lifecycle. Original Rangoon explanatory diagram: concepts for editorial review, not measured results, an official legal determination, vendor interface or event photograph.",
    "credit": "Rangoon / Hypler",
    "source_url": "https://rangoon.ai/brand/",
    "rights_url": "https://rangoon.ai/terms/#media-use",
    "rights_note": "Hypler artwork permissions do not grant rights to third-party marks or imply endorsement."
  },
  "sections": [
    {
      "id": "memo-event",
      "url": "https://rangoon.ai/insights/archive-omb-m-25-22-ai-procurement-april-2025/#memo-event",
      "heading": "M-25-22 addressed the acquisition path",
      "paragraphs": [
        "OMB Memorandum M-25-22, “Driving Efficient Acquisition of Artificial Intelligence in Government,” is dated April 3, 2025. Its text says it provides guidance intended to improve agencies’ ability to acquire AI responsibly, referencing Executive Order 14179 and M-25-21. That cross-reference matters: M-25-22 does not erase the separate use-governance questions described in M-25-21.",
        "The contemporaneous documents therefore support two related but distinct records. Procurement asks what is being bought or accessed, on what terms and with which performance, security and data conditions. Use governance asks whether a defined mission use is acceptable and controlled. One supplier agreement may support many possible uses, and one approved use may need no new purchase."
      ],
      "bullets": [],
      "source_ids": [
        1,
        2
      ]
    },
    {
      "id": "requirements",
      "url": "https://rangoon.ai/insights/archive-omb-m-25-22-ai-procurement-april-2025/#requirements",
      "heading": "Acquisition begins with measurable requirements",
      "paragraphs": [
        "A useful AI solicitation does not ask only for a model family, parameter count or general capability claim. It states workloads, data categories, integration boundaries, latency and availability targets, evaluation fixtures, accessibility needs, incident response, audit access, retention, subcontractor terms and termination or export path. Those requirements give bidders something testable and give the buyer a basis for acceptance.",
        "Separate supplier assertions from evidence. A provider’s model card, certification or demo may be useful input, but it is not automatically evidence that a system meets a mission-specific interface, security or accuracy requirement. Require a dated configuration, fixed test materials where permitted, observed outputs and a record of exceptions. This analysis does not prescribe any federal contract clause."
      ],
      "bullets": [],
      "source_ids": [
        1,
        2
      ]
    },
    {
      "id": "lifecycle-receipt",
      "url": "https://rangoon.ai/insights/archive-omb-m-25-22-ai-procurement-april-2025/#lifecycle-receipt",
      "heading": "Analysis: preserve evidence from award through operation",
      "paragraphs": [
        "The acquisition record should survive award. Store the evaluated offer, pricing basis, named services, model and region assumptions, security commitments, change-notice terms and acceptance results. At operation, log which purchased capability was enabled, which configuration ran and whether a material change required new testing or approval. A contract can be valid while a changed service no longer matches the evaluated product.",
        "Plan exit before a service becomes mission-critical. Identify portable data formats, retrieval time, deletion evidence, transition assistance and the owner who decides when to stop. These are ordinary lifecycle questions, not a prediction that a particular supplier will fail. Their value is preserving agency options when price, performance, security or policy conditions change."
      ],
      "bullets": [],
      "source_ids": [
        1,
        2
      ]
    },
    {
      "id": "historical-boundary",
      "url": "https://rangoon.ai/insights/archive-omb-m-25-22-ai-procurement-april-2025/#historical-boundary",
      "heading": "Analysis: award is not deployment, and memo is not award",
      "paragraphs": [
        "The April 2025 memorandum is evidence of OMB guidance; it is not evidence of a contract award, a deployed model, price, security assessment or service result. Likewise, an award notice does not establish that the system passed acceptance testing or was used for a particular decision. Historical reporting should keep those states distinct.",
        "Readers with live procurement questions should consult current official acquisition authorities, agency procedures and qualified professionals. This article preserves a dated policy event and offers a recordkeeping pattern; it does not provide procurement or legal advice."
      ],
      "bullets": [],
      "source_ids": [
        1,
        2
      ]
    }
  ],
  "project_context": null,
  "related_links": [],
  "permissions_url": "https://rangoon.ai/permissions.txt",
  "content_hash": "sha256:389446f24672d432acf4d4d22d112f197412bf8bb28abe711209af995039084d"
}