Preventing Woke AI in the Federal Government

7/23/2025

Executive Orders

View Original

Action Summary

Purpose: Establish a federal policy to ensure AI models, particularly large language models (LLMs), produce reliable, fact-based outputs by avoiding the inclusion of ideological biases—specifically those related to diversity, equity, and inclusion (DEI)—in procurement decisions.
Definitions: Clarifies key terms including:
- Agency: Executive departments, military departments, or independent establishments.
- Agency Head: The top official(s) within an agency.
- LLM: Large language models enabling natural-language responses from diverse datasets.
- National Security System: As defined in applicable law.
Unbiased AI Principles: Outlines two core mandates for LLMs:
- Truth-seeking: Models must provide accurate, historically faithful, and evidence-based information while noting uncertainties.
- Ideological Neutrality: Models must remain nonpartisan and refrain from skewing outputs to favor ideological positions like DEI.
Implementation: Details the process for integrating these principles into federal procurement:
- Within 120 days, senior officials (including OMB and federal procurement leaders) will issue guidance to agencies.
- Guidance will address technical limitations, transparency requirements, and allow for vendor flexibility.
- New and modified contracts must enforce compliance with the Unbiased AI Principles, including provisions for decommissioning noncompliant models.
- Special considerations are provided for LLM use in national security systems.
General Provisions: Specifies that the order:
- Does not impair the authority of executive departments or agency heads.
- Is subject to appropriations and applicable law.
- Creates no enforceable rights for third parties.
- Charges publication costs to the General Services Administration.

Risks & Considerations

The Executive Order emphasizes the need for AI models to be free from ideological biases, particularly those related to diversity, equity, and inclusion (DEI). This could impact Vanderbilt University’s research and development in AI, especially if current projects incorporate DEI principles.
There is a risk that the order could limit the scope of AI research and development at Vanderbilt, as it may require adjustments to ensure compliance with the new federal procurement standards for AI models.
The focus on “truth-seeking” and “ideological neutrality” may necessitate changes in the training and development of AI models, potentially increasing costs and requiring additional resources to align with federal guidelines.
Vanderbilt may need to reassess its partnerships with AI vendors and collaborators to ensure that their AI models meet the new federal standards, which could affect ongoing and future projects.
The order could influence the university’s approach to AI ethics and governance, requiring updates to policies and practices to align with the federal emphasis on unbiased AI.

Impacted Programs

Vanderbilt’s Data Science Institute may need to adjust its research focus and methodologies to ensure compliance with the new federal AI standards, potentially affecting ongoing projects and collaborations.
The School of Engineering could see changes in its AI curriculum and research initiatives, as the order may necessitate a shift towards developing AI models that prioritize truth-seeking and ideological neutrality.
The Office of Research might need to implement new compliance measures and oversight to ensure that AI-related research aligns with federal procurement requirements.
Vanderbilt’s partnerships with tech companies may require reevaluation to ensure that collaborative AI projects adhere to the new federal guidelines.

Financial Impact

The order could lead to increased costs for AI research and development, as Vanderbilt may need to invest in new technologies and training to comply with the federal standards for unbiased AI.
There may be potential funding opportunities for projects that align with the federal emphasis on trustworthy AI, but securing these funds could require significant adjustments to current research practices.
Vanderbilt might face financial risks if existing AI projects are found to be non-compliant with the new standards, potentially leading to contract terminations or the need for costly modifications.
The university’s ability to attract federal grants for AI research could be impacted by its compliance with the new order, necessitating strategic adjustments in grant applications and project proposals.

Relevance Score: 4 (The order presents a need for potential major changes or transformations of programs.)

Key Actions

Vanderbilt’s AI Research and Development Teams should evaluate the implications of the executive order on their AI projects, ensuring that any AI models developed or used align with the Unbiased AI Principles of truth-seeking and ideological neutrality. This may involve revisiting current AI training datasets and methodologies to ensure compliance.
The Office of Federal Relations should monitor the guidance issued by the Office of Management and Budget (OMB) and other relevant federal bodies to understand how these new requirements might affect federal funding and procurement opportunities for AI-related projects.
Vanderbilt’s Legal and Compliance Teams should review existing contracts involving AI technologies to ensure they meet the new federal requirements, and prepare to renegotiate terms if necessary to avoid potential decommissioning costs.
The Department of Political Science could conduct research on the broader societal impacts of the executive order, particularly how it might influence the development and deployment of AI technologies in educational and research settings.
Vanderbilt’s Center for Technology and Society should explore opportunities to engage in public discourse and policy advocacy regarding the balance between AI innovation and regulatory compliance, positioning Vanderbilt as a thought leader in this evolving field.

Opportunities

The executive order presents an opportunity for Vanderbilt’s AI Research Initiatives to lead in developing AI models that prioritize truthfulness and neutrality, potentially setting industry standards and attracting federal partnerships.
By aligning with the Unbiased AI Principles, Vanderbilt can enhance its reputation as a center for ethical AI research, attracting students, faculty, and funding interested in advancing trustworthy AI technologies.
The focus on transparency and disclosure in AI development offers an opportunity for Vanderbilt’s Data Science Institute to innovate in creating transparent AI systems, potentially leading to new research collaborations and funding opportunities.

Relevance Score: 4 (The order necessitates major process changes in AI development and compliance to align with federal procurement standards.)

Average Relevance Score: 3.8

Timeline for Implementation

Within 120 days of July 23, 2025: The Director of the Office of Management and Budget shall issue guidance to agencies for implementing the Unbiased AI Principles.
Within 90 days after the issuance of the OMB guidance: Each agency head must adopt procedures to ensure that LLMs procured by the agency comply with the Unbiased AI Principles.

Determination: The shortest timeline specified is 90 days after the OMB guidance, which falls under the 90-179 day range.

Relevance Score: 2

Impacted Government Organizations

Office of Management and Budget (OMB): The Director is tasked with issuing guidance to agencies for implementing the Unbiased AI Principles and coordinating with other designated offices.
Administrator for Federal Procurement Policy: In consultation with the OMB Director, this role plays a key part in establishing procurement guidelines and contract modifications for AI systems within Federal agencies.
General Services Administration (GSA): Charged with responsibilities related to contracting terms, including decommissioning cost provisions for noncompliant AI models, and handling the publication costs of the guidance.
Office of Science and Technology Policy (OSTP): Consulted to provide technical insight as part of the guidance process for ensuring that large language models adhere to the Unbiased AI Principles.
Federal Agency Heads: All agency leaders are mandated to ensure that AI procurement complies with the standards set forth in the order, impacting the entire landscape of Federal procurement and policy implementation.

Relevance Score: 5 (The directive applies across the entire Federal government’s procurement and policy framework.)

Responsible Officials

Director, Office of Management and Budget (OMB) – Tasked with issuing guidance to agencies for implementing the Unbiased AI Principles within 120 days.
Administrator for Federal Procurement Policy – Consulted by the OMB Director to help shape the procurement guidance for unbiased AI.
Administrator of General Services – Works in consultation with the OMB and other officials to revise contracting terms and support the procurement process.
Director, Office of Science and Technology Policy – Provides technical and policy input in the consultation process for issuing guidance.
Agency Heads – Responsible for ensuring that all Federal contracts for large language models comply with the Unbiased AI Principles, including contract modifications and compliance procedures.

Relevance Score: 4 (Directives significantly affect agency heads and key agency directors involved in Federal procurement and oversight of AI technology).