GAO's Finding
In its November 18, 2026 report, Strategic Planning: FDA Could Further Strengthen Robust IT Modernization Plan, the Government Accountability Office examined the Food and Drug Administration's approach to information technology modernization.
FDA has developed a strategic planning framework supporting 17 major IT investments. GAO found that the agency's plans fully addressed six of eight leading strategic-planning practices, including alignment with agency missions, performance measures, and defined responsibilities.
The remaining weaknesses are significant.
GAO found that FDA had not fully documented how it would track mitigation of key modernization challenges or demonstrate the relationship between its major IT investments and strategic goals. In its review of five major investments, GAO also found that FDA had not demonstrated how performance metrics were used to manage and improve one investment.
GAO issued three recommendations addressing these weaknesses. The Department of Health and Human Services agreed with all three.
These findings do not establish that FDA's modernization has failed. They establish that important relationships between strategic intent, investments, risks, and measured performance require stronger documentation.
The Perceived Organizational Information Continuity Challenge
GovKM identifies three related continuity challenges.
1. Strategic objectives and investment decisions must remain connected.
An IT strategic plan establishes objectives. Individual investments produce decisions, expenditures, implementation records, and performance evidence. If the relationships between those records are not maintained, an agency may retain the documents without being able to reconstruct why a particular investment was authorized, which objective it supports, or whether subsequent changes remain justified.
Consequence for AI trustworthiness: An AI assistant may retrieve the investment justification and the strategic plan independently without establishing whether they remain aligned. It could present an outdated justification as current or infer a relationship unsupported by the authoritative record.
This is a potential AI-readiness risk identified by GovKM, not an AI failure reported by GAO.
2. Modernization risks must remain connected to corrective action.
Recording a challenge is insufficient if its status, responsible authority, mitigation decision, and supporting evidence cannot be reconstructed over time.
Personnel changes, reorganizations, system migrations, and revisions to project plans can separate the current risk assessment from the decisions that produced it.
Consequence for AI trustworthiness: An AI system may repeat an obsolete risk assessment, overlook a corrective action, or present an unresolved issue as closed. Without traceable status and authority, neither the user nor the AI system can reliably establish which information governs the current situation.
3. Performance measurements must remain connected to management decisions.
Performance metrics have limited governance value when the organization cannot demonstrate how measured results informed decisions, corrective actions, and subsequent outcomes.
Consequence for AI trustworthiness: An AI assistant might accurately report a metric while incorrectly implying that it demonstrates program success. A trustworthy assessment requires the metric's definition, reporting period, authoritative source, applicable target, and relationship to the management decision.
The broader issue is not whether an organization possesses sufficient information. It is whether the information environment preserves the relationships necessary to establish what was known, what was authorized, what changed, and what the evidence demonstrates.
GovKM's Recommended Solution
GovKM recommends an Organizational Information Continuity architecture that connects strategic objectives, authorized investments, implementation decisions, risks, corrective actions, performance measures, and supporting records within a governed information environment.
This is a proposed implementation method, not a claim that GAO prescribed GovKM's architecture or that FDA has adopted it.
Establish authoritative records and accountable ownership.
Each major investment should have identifiable authoritative records for its objectives, approval decisions, scope, funding, responsible officials, milestones, risks, and performance measures.
Under 44 U.S.C. § 3101, federal agency heads must make and preserve records adequately documenting agency functions, policies, decisions, procedures, and essential transactions.
36 CFR § 1222.22 further specifies the records needed to document agency business, including the formulation and execution of basic policies and decisions.
GovKM's recommended architecture translates these obligations into practical relationships among responsible officials, decisions, records, and supporting evidence.
Maintain traceability between strategic goals and investments.
For every major investment, establish and maintain a documented connection to the strategic objectives it supports, the authority under which it proceeds, the resources committed, and the outcomes against which it will be evaluated.
This directly addresses the investment-alignment issue identified in GAO-27-108743.
OMB Circular A-130 provides federal policy for managing information as a strategic resource, including responsibilities for information resources management, governance, and investment oversight.
GovKM recommends implementing those responsibilities through explicit, reviewable links between the planning, investment, and operational evidence. The precise controls must be assessed against applicable agency requirements.
Preserve risk and corrective-action histories.
Each material modernization challenge should have an accountable owner, dated assessment, mitigation decision, supporting evidence, current status, and documented closure criteria.
Changes should preserve the prior operative state rather than silently replacing it.
Federal recordkeeping requirements in 36 CFR § 1222.26 address responsible offices, record systems, electronic-record metadata, preservation, and retrieval. These requirements provide a regulatory foundation for managing records needed to reconstruct agency actions.
GovKM's proposed continuity controls extend that discipline into a traceable operating method for investment and risk management.
Connect performance evidence to decisions.
Performance measures should identify their source, definition, reporting period, responsible official, target, and related investment.
Management decisions should reference the performance evidence actually considered. Corrective actions should remain connected to their triggering findings and subsequent verification.
This allows reviewers to distinguish between a metric that was merely collected and one that demonstrably informed management.
Prepare the information environment for trustworthy AI.
AI access should respect authoritative sources, approved permissions, record status, effective dates, and the distinction between current and superseded information.
An AI-generated answer about an investment should be verifiable against the applicable strategic objective, authorized decision, performance evidence, and current risk status.
Where evidence is missing or conflicting, the system should expose that limitation rather than manufacture certainty.
GovKM's proposed OrgAI approach is intended to help maintain these governed information relationships over time. Its proposed role does not replace agency decision authority, federal records responsibilities, or independent compliance verification.
How an Agency Would Verify the Solution
An agency could test the architecture by selecting a major IT investment and asking an independent reviewer to reconstruct its history.
The reviewer should be able to establish which strategic objective authorized the investment, which officials made consequential decisions, what risks were known, what evidence supported mitigation, how performance changed, and whether subsequent decisions reflected those results.
The same test can be applied to an AI assistant. Its answers should identify the authoritative supporting records, distinguish current from superseded information, respect access restrictions, and disclose material evidence gaps.
Successful testing would demonstrate stronger information traceability and support for justified reliance on AI outputs. It would not, by itself, establish full regulatory compliance or guarantee AI accuracy.
Beyond FDA
The underlying challenge extends beyond federal IT modernization.
A government agency migrating legacy records into Microsoft 365, for example, must preserve more than the documents themselves. It must retain the authority, classification, provenance, retention obligations, and decision context needed to interpret those documents after migration.
Private organizations face comparable information-continuity problems, although their legal obligations and applicable standards differ.
The method is transferable. The governing requirements must be determined separately.
GovKM's Role
An organization should be able to explain not only what technology it purchased or deployed, but why it made that decision, what evidence supported it, who had authority, and whether the expected result was achieved.
That requirement becomes more consequential when AI systems are expected to interpret organizational information and support future decisions.
GovKM helps organizations assess, prepare, implement, and validate governed Organizational Information Continuity environments that preserve these relationships and strengthen AI readiness.
For agencies and organizations evaluating modernization, records governance, or AI adoption, contact GovKM to discuss a continuity assessment and evidence-based implementation approach.



