FEDERAL FUNDING & OVERSIGHT

House Bill Would Restore FCC's Disbanded Equity Council

• Menendez, Matsui, Barragán, and Carter Introduce Legislation Compelling FCC to Re-Establish Communications Equity and Diversity Council 

On March 16, 2026, Representatives Rob Menendez (NJ-08), Doris Matsui (CA-07), Nanette Barragán (CA-44), and Troy Carter, Sr. (LA-02) introduced H.R. 7943, the Communications Equity and Diversity Council Act, which would require the Federal Communications Commission to restore the Communications Equity and Diversity Council (CEDC).

The legislation responds to FCC Chair Brendan Carr’s January 2025 decision to disband the CEDC, an advisory body that had operated since 2003 under both Democratic and Republican administrations. The council previously advised the FCC on equitable access to communications services, including broadband deployment in underserved communities, digital inclusion, minority-owned business participation, and workforce development.

The bill would codify the CEDC in statute, requiring a membership of no fewer than 30 individuals and limiting the agency’s ability to eliminate the council through administrative action. The proposal draws a parallel to other FCC advisory bodies that have been codified by Congress.

The legislation is supported by a coalition of civil rights, consumer advocacy, and technology policy organizations. It has been introduced and has not yet advanced through the legislative process.

See Also: FCC Investigates Corporate DEI Programs (Issue 7); FCC Conditions Verizon-Frontier Merger Approval on DEI Dismantling (Issue 13); AT&T Refuses to Follow Verizon (Issue 14); FCC Expands EEO Audits to Include DEI Program Scrutiny (Issue 26); AT&T Commits to Ending DEI Programs Under FCC Pressure (Issue 41)

FEDERAL FUNDING & OVERSIGHT

Depositions Reveal AI-Driven Review Used to Terminate Federal DEI Grants

• When DOGE Unleashed ChatGPT on the Humanities

• Major Update in our NEH Lawsuit - AHA
  

• The Authors Guild v. NEH - Deposition of Justin Fox

• The Authors Guild v. NEH - Deposition of Nathan Cavanaugh

• The Authors Guild v. NEH - Deposition of Michael McDonald 

OVERVIEW

Discovery materials filed on March 6, 2026, in consolidated litigation brought by the American Council of Learned Societies, the American Historical Association, the Modern Language Association, and the Authors Guild provide the clearest record to date of how National Endowment for the Humanities grants were identified for termination during the Trump administration. The filing, submitted in support of plaintiffs’ motion for summary judgment in the Southern District of New York, argues that Department of Government Efficiency staff assumed control over key grant-termination decisions and carried out cuts that violated the First Amendment, exceeded statutory authority, and in many cases violated the equal protection guarantee of the Fifth Amendment.

According to the filing and supporting deposition testimony, former DOGE staffers Justin Fox and Nathan Cavanaugh used keyword searches and ChatGPT-generated summaries to identify grants as involving DEI. The record states that grants were screened for terms related to race, gender, sexuality, and other protected characteristics, and that Fox used prompts asking whether a grant involved DEI “from the perspective of someone looking to identify DEI grants,” with short responses used to generate a “DEI rationale.” Fox then combined those AI-generated assessments with NEH risk designations in a spreadsheet cataloguing 1,057 grants as involving DEI.

The court record also includes deposition testimony that DOGE staff did not apply safeguards to assess whether the AI analysis could classify grants on the basis of protected characteristics. In one exchange, Fox testified that he did nothing to ensure ChatGPT’s DEI analysis would not discriminate on the basis of race and stated that it did not matter whether it discriminated on the basis of sex. 

The filing also cites examples of grants classified as DEI because they addressed Black civil rights, Jewish writers and the Holocaust, or other projects involving race, ethnicity, religion, gender, or sexual orientation.

Plaintiffs assert that these grants were canceled through an ideologically driven process that discriminated on the basis of race, ethnicity, gender, and other characteristics, and that DOGE staff lacked lawful authority to make those determinations. The case remains pending, and the March 6 summary judgment filing and supporting discovery materials are now part of the court record.

LEGAL INTERPRETATION

This case centers on whether federal officials acted within their statutory and constitutional authority in terminating congressionally funded grants.

Federal grant programs, such as those administered by the National Endowment for the Humanities, are governed by congressional appropriations and enabling statutes. Agencies are required to administer those funds consistent with legislative intent. Plaintiffs allege that grants were terminated based on criteria not authorized by statute and that individuals outside the agency exercised decision-making authority, raising questions about whether those actions exceeded lawful authority.

Under the Administrative Procedure Act, agency actions must be reasoned and not arbitrary or capricious. Courts reviewing the record will assess whether the grant terminations were supported by a rational explanation and followed required decision-making processes. Deposition testimony describing the use of keyword searches and AI-generated summaries as part of the review process will be considered as part of that analysis.

The claims also include constitutional challenges. Plaintiffs argue that the grant terminations violated the First Amendment by targeting projects based on subject matter and viewpoint, and violated the equal protection guarantee of the Fifth Amendment by discriminating on the basis of race, ethnicity, gender, and other protected characteristics.

In addition, plaintiffs raise claims related to federal records requirements, including the use of non-government communication platforms in connection with the grant review process, which may implicate obligations under the Federal Records Act.

BRIDGE POV

Important programs were eviscerated with a single ChatGPT prompt.

They were not reviewed. They were not debated. They were not evaluated on merit. They were reduced to keywords and labeled for elimination.

Congressionally funded programs tied to history, culture, and scholarship were filtered through a process that prioritized speed over expertise and output over understanding.

This is exactly how risk shows up inside companies.

When pressure rises and scrutiny increases, organizations look for shortcuts. They rely on signals. They move faster than their systems are designed to handle. Increasingly, they turn to AI to do the sorting.

But AI reflects the question it is asked. If the question lacks clarity, the outcome will, too.

What the record shows is a breakdown in decision-making discipline. No clear definitions. No subject-matter expertise. No safeguards to ensure accuracy or fairness.

Inclusion is not a keyword. It is a capability. It requires judgment, context, and precision.

The companies that will lead through this moment are the ones that stay rigorous, define their standards, and ensure that decisions of consequence are made with accountability, not automation.

ACTIONABLE STRATEGIES

1. Define decision standards before tools are applied: Do not allow keywords, labels, or external pressure to determine outcomes. Establish clear internal definitions for inclusion, compliance, and business impact before any evaluation process begins.
   
   
2. Put guardrails around how AI is used: Audit where AI is being used to evaluate programs, investments, or risk. Ensure outputs are treated as inputs, not conclusions, and require review by qualified leaders in high-impact decisions.
   
   
3. Require accountability for consequential decisions: Decisions that affect funding, people, or reputation must have clear ownership. Document who is making the decision and the criteria being applied to ensure accountability and consistency.
   
   

See also: USDA Keyword Purge Targets "Diversity" and "Climate" Grants (Issue 39); Fourth Circuit Declines Facial Challenge to Anti-DEI Orders, Keeps Door Open for Future Legal Action (Issue 52)

COURTS & LITIGATION

17 States Sue to Block Federal Admissions Data Requirement

• Democratic-led states sue over Trump demand for colleges to supply race data 

OVERVIEW

On March 11, 2026, a coalition of 17 Democratic-led states filed suit in the U.S. District Court for the District of Massachusetts challenging a Department of Education directive requiring colleges and universities to submit detailed admissions data disaggregated by race and other demographic and academic factors.

The requirement is part of an expansion of federal data collection tied to the Integrated Postsecondary Education Data System (IPEDS), which institutions must complete to participate in federal student aid programs. The directive calls for the submission of multiple years of admissions-related data, including information on race, sex, academic metrics, and other applicant characteristics.

The data collection initiative implements an August 7, 2025 presidential memorandum directing the Department of Education to expand reporting to assess institutional compliance with the Supreme Court’s 2023 decision in Students for Fair Admissions v. Harvard. The Office of Management and Budget approved the data collection in December 2025.

The states allege that the Department exceeded its statutory authority and imposed new requirements without proper administrative process. They are seeking to block enforcement of the directive.

On March 13, 2026, the court issued a temporary restraining order extending the data submission deadline while it considers the states’ request for broader relief. The case remains ongoing.

LEGAL INTERPRETATION

This case centers on the scope of federal authority to require data collection from institutions that participate in federal student aid programs.

Colleges and universities receiving Title IV funding are subject to reporting requirements administered through the Department of Education, including data submitted through the Integrated Postsecondary Education Data System. The states challenge whether the Department can expand those requirements to mandate additional admissions data without statutory authorization.

The claims also raise issues under the Administrative Procedure Act. Plaintiffs allege that the Department imposed new obligations unlawfully and failed to comply with legal requirements governing federal data collection, including privacy impact assessment obligations under the E-Government Act of 2002 and requirements under the Paperwork Reduction Act.

The directive is also tied to federal enforcement of the Supreme Court’s 2023 decision in Students for Fair Admissions v. Harvard. As a result, the case may clarify how far federal agencies can go in requiring institutions to produce admissions data to assess compliance with that ruling.

BRIDGE POV

The federal government is directing colleges and universities to turn over detailed admissions data tied to race and other student characteristics, on a compressed timeline and with significant consequences for getting it wrong.

When requirements move this quickly, organizations default to reaction. Collect more. Report more. Move faster.

But more data does not lead to better decisions.

If there is no clarity on why data is being collected or how it will be used, risk builds quickly, including legal exposure, reputational risk, and operational strain.

The organizations that handle this well will not chase every new requirement. They will be disciplined about what they collect, why they collect it, and how it connects to actual business and compliance needs. They will align legal, data, and leadership teams before acting, not after.

Because in this environment, how you respond matters as much as what you report.

ACTIONABLE STRATEGIES

1. Be clear on why you are collecting data: Do not collect or report sensitive data without a defined legal basis and business purpose. Align internally on what is required, what is optional, and what you are choosing to do.
   
   
2. Put structure around how sensitive data is used: Set clear rules for how demographic data is collected, accessed, and applied. Limit use to defined purposes and ensure oversight for anything tied to protected characteristics.
   
   
3. Slow down reactive responses: When new requirements emerge, resist the instinct to over-collect or over-report. Validate what is actually required, align your teams, and move with intention.

See also: Trump Administration Adopts New Reporting Requirement for Colleges and Universities to Ascertain Whether Race Is Being Used in Admissions (Issue 25)

COURTS & LITIGATION