Purported DOGE Contract Review Tool Cited in Reports of AI-Driven Misjudgments in VA Budget Cuts

A purported AI tool using LLMs was deployed to classify Veterans Affairs contracts as expendable based on limited text and simplified criteria. The system allegedly produced hallucinated values and flagged critical healthcare and research services for cancellation. Reportedly, at least two dozen flagged contracts were later terminated.

Risk classification

• Primary risk domain: 7 AI system safety, failures, & limitations
• Primary risk subdomain: 7.3 Lack of capability or robustness

The AI system failed to perform reliably, producing glaring errors and hallucinations regarding contract values and essential services due to technical limitations.

Additional risk subdomains

• 5.1 Overreliance and unsafe use: DOGE and VA officials over-relied on a hastily built, unvalidated AI tool to make critical budgetary decisions affecting veteran healthcare.

Causal factors

• Entity: AI
• Intent: Unintentional
• Timing: Post-deployment

The risk was caused by the AI system's technical failures and hallucinations after it was deployed to analyze VA contracts.

EU AI Act risk tier

• Risk tier: 2 High Risk

High Risk: The AI system was used in healthcare administration and public services (Veterans Affairs), directly impacting access to essential healthcare services and medical research, which has significant implications for safety and fundamental rights.

AI system and alleged parties

• AI system: DOGE AI tool
• AI purpose: Resource Allocation; Value Estimation
• Behaviour type: Assistant
• Alleged developer: Sahil Lavingia, DOGE, Department of Government Efficiency
• Alleged deployer: Sahil Lavingia, DOGE, Department of Government Efficiency
• Alleged harmed parties: Veterans receiving care through the VA, Veterans, VA contractors, VA clinical and research staff, Department of Veterans Affairs (VA)

Harm severity

Severity ratings are based on this scale.

Highest direct severity in any category: Minor. Severity is scored from Negligible to Catastrophic in each harm category, for harm the reports describe as caused directly or indirectly by the AI system.

• Physical: direct Negligible, indirect Minor
• Infrastructure: direct Negligible, indirect Negligible
• Property: direct Negligible, indirect Negligible
• Financial: direct Minor, indirect Negligible
• Environmental: direct Negligible, indirect Negligible
• Malicious content: direct Negligible, indirect Negligible
• Differential treatment: direct Negligible, indirect Negligible
• Civil rights: direct Negligible, indirect Negligible
• Democracy: direct Negligible, indirect Negligible
• Privacy: direct Negligible, indirect Negligible
• Psychological: direct Negligible, indirect Negligible
• Epistemic: direct Minor, indirect Minor
• Child sexual exploitation and abuse: direct Negligible, indirect Negligible

Physical

Reported: The report explicitly describes indirect physical harm to veterans due to the cancellation of critical healthcare and research contracts.

Indirectly caused: The cancellation of at least 24 contracts, including those for cancer treatment gene sequencing and blood sample analysis, directly jeopardized the quality of care provided to veterans.

Inferred additional harm: It is highly likely that dozens of veterans experienced delayed, degraded, or interrupted medical treatments, potentially leading to worsened health outcomes or physical suffering.

Financial

Reported: The report explicitly describes financial losses resulting from the cancellation of contracts, though the total value is not fully quantified.

Directly caused: At least 24 contracts were canceled, resulting in direct financial losses to the affected contractors, with some individual contracts valued up to $35,000 or more.

Inferred additional harm: With over 2,000 contracts flagged for cancellation, the total financial loss to contractors and the VA's research ecosystem is likely in the millions of dollars.

Epistemic

Reported: The report explicitly describes epistemic harm in the form of AI-driven hallucinations and misinformation regarding contract values.

Directly caused: The AI tool hallucinated contract values, frequently misreading them and inflating their value, such as concluding contracts were worth $34 million when they were worth $35,000.

Indirectly caused: These fabricated values and incorrect classifications misled decision-makers, leading to the cancellation of essential contracts.

People affected

• Occurrences reported: 1
• People reportedly harmed: 50
• People reportedly exposed: 2000

Potential causes

Management

• Speed Prioritized Over Safety: Management pushed for rapid cuts without evaluating operational risks.
• Inadequate Risk Assessment: Deployed unvalidated AI tools for critical budgetary decisions.
• Opaque Decision-Making: Leadership failed to provide transparency on final cancellation choices.

Technology

• Outdated LLM Models: Used inexpensive, older models incapable of solving complex tasks.
• Model Hallucinations: AI hallucinated contract sizes, inflating value of small contracts.
• Flawed Prompt Engineering: Prompts lacked essential context about VA operations and legal mandates.

Data Inputs

• Truncated Input Text: Analyzed only the first 2500 words, missing crucial contract details.
• Inaccurate Value Extraction: AI selected wrong dollar values when multiple figures were present.
• Ignored Public Databases: Failed to verify contract data against readily available public sources.

Human Factors

• Lack of AI Expertise: Developer had no formal training in artificial intelligence.
• No Healthcare Background: Developer had no experience in healthcare or government procurement.
• Cognitive Overload on Staff: VA employees faced extreme pressure to justify contracts in hours.

Process and Methods

• Rushed Software Development: The tool was built in a single day using AI-assisted coding.
• Restricted Appeal Process: Staff restricted to 255-character responses to defend contracts.
• Opaque Review System: VA staff could not see who made final decisions to cancel contracts.

Regulatory Environment

• Aggressive Federal Timeline: Executive order required a complete contract review within 30 days.

Information quality

• Classification confidence: High
• Reason for confidence: The reports from ProPublica and Seifeur provide highly detailed, consistent accounts of the AI tool's development, its specific technical flaws (hallucinations, 2500-word limit), the developer's background and admissions, and the specific real-world impacts (the cancellation of 24 contracts including cancer research). The role of the AI is explicitly documented.
• Ambiguities identified: The exact number of total contracts canceled beyond the 24 identified by ProPublica remains unclear due to the opaque nature of the VA's decision-making process.
• Alternative interpretations: None. The tool's failure is clearly a mix of technical limitations (7.3) and human overreliance/rushed deployment (5.1).

The deployment of an unvalidated, flawed AI tool by DOGE to evaluate VA contracts led to severe hallucinations and the erroneous cancellation of critical veteran healthcare and research services. While causing direct harm to veteran care and procurement processes, the incident remains a domestic administrative failure with minor national security implications.

• Overall national security impact: Minor
• Scope: Single nation
• Primary target: United States
• Alleged perpetrator: Department of Government Efficiency (DOGE)

Threat characteristics

• Imminence: Near-term. The situation represents an active disruption of veteran healthcare services requiring near-term administrative remediation.
• Autonomy: Human-supervised. The AI flagged contracts, and although human review was highly constrained (hours or 255 characters), humans ultimately executed the cancellations.
• Novelty: Evolved capability. Represents a significant escalation in the deployment of unvalidated LLMs for critical government administrative and budgetary decision-making.

Impact by dimension

• Physical security: Minor. The AI tool's failure led to the cancellation of 24 critical healthcare and research contracts, impacting veteran care, but this remains an administrative issue manageable within standard government procedures.
• Information security: Negligible. No intelligence compromise, classified data theft, or foreign information warfare operations occurred.
• Sovereignty: Minor. The incident represents a domestic administrative failure that disrupted Veterans Affairs operations and decision-making, but it does not threaten state authority or constitutional processes.
• Economic security: Minor. Financial losses occurred for affected contractors due to canceled agreements, but there was no threat to strategic industries or national economic stability.
• Societal stability: Minor. The cancellation of medical and research contracts negatively impacts a specific subpopulation (veterans), but it does not represent systematic discrimination or a threat to overall social cohesion.