The coalition letter of 9 September 2026 is a two-page letter to President Donald Trump, signed by twenty-eight organizations and hosted by Americans for Responsible Innovation, asking the Administration "to publicly release the voluntary AI framework supporting the White House's June 2 Executive Order." The letter is framed as supportive of the Administration's objectives — it opens by appreciating "the Administration's efforts to advance artificial intelligence (AI) systems in an innovative and safe manner" — and does not oppose the review process itself. Its objection is to the process being confidential.
The ask
The letter states that "the federal government's new framework for reviewing frontier AI systems must be public and should be developed in a transparent manner." It allows that "there may be legitimate reasons to withhold some portions for the sake of confidentiality or national security where it's necessary," but asserts that "the government has an unequivocal duty to disclose information about the framework's standards, requirements, structure, and operational details" — four enumerated items. The letter supplies no test for when withholding is necessary, no process for determining it and no reviewer, so it should not be read as proposing a confidentiality standard.
The arguments
The letter asserts three distinct harms from a confidential framework.
Public accountability. It characterises the framework as "[i]mplementing a confidential process where the federal government obtains priority access to frontier AI models, technology with exponential power," and states that through the framework the federal government "will obtain initial access and visibility into the most transformative technologies we have ever seen. There rests the power to deny or restrict access for the general public." It argues the government "has a fundamental responsibility to share the parameters of the process for reviewing frontier AI models prior to wider deployment," and that a lack of accountability over the governance of such a technology "would be inconsistent with democratic values and likely lead to a loss of public trust."
Competitive effect. The letter states that a confidential framework "disadvantages the work of smaller and open-source labs," and that "[a] framework shared only with the largest AI companies may unduly influence market structure, enable regulatory capture, and entrench the use of models created by the largest AI labs."
The congressional process. It argues that as Congress considers AI regulation, "it is all the more important that policymakers have full access to the Administration's framework as they work to codify guardrails on frontier AI," and that the Administration "has an opportunity to help lead the conversation on AI policy happening now in Congress, but only by transparently sharing its own vision for AI oversight."
The letter closes by tying disclosure to the competition with China: "[w]ith more public discourse we can ensure the government is successful in its goal to maintain its lead over China in the AI race and to advance innovative and safe AI frontier models."
Signatories
Twenty-eight organizations sign, in the order given: Americans for Responsible Innovation; Center for Democracy and Technology; Advocacy for Principled Action in Government; The Alliance for Secure AI; Americans for Prosperity; Autistic Women & Nonbinary Network; Center for American Progress; Center for Oil & Gas Organizing; Common Cause; Consumer Federation of America; Demand Progress; Design It For Us; Fight for the Future; Free Press Action; Government Information Watch; Inclusive Abundance Action; National Consumers League; National Fair Housing Alliance; NAVA (National Association of Voice Actors); New America's Open Technology Institute; Project On Government Oversight; Public Citizen; Public Good Group; Public Knowledge; Restore The Fourth - Action; R Street Institute; The Tech Oversight Project; and X-Lab.
The list is not confined to AI-policy organizations. It spans consumer groups, transparency and good-government groups, civil-liberties and digital-rights organizations, a fair-housing organization, a disability-community organization, a voice actors' association and a labor-organizing group. Two signatories answer to the "free-market" description used in press accounts that did not name one: the R Street Institute describes itself as a free-market think tank, and Americans for Prosperity is a free-market advocacy organization. Which one the coverage meant cannot be established from the letter.
The letter is copied to Commerce Secretary Howard Lutnick, National Cyber Director Sean Cairncross, Office of Science and Technology Policy Director Michael Kratsios, Senate Majority Leader John Thune, Senate Minority Leader Chuck Schumer, Speaker of the House Mike Johnson, and House Minority Leader Hakeem Jeffries.
What the letter does not supply
The letter is written from outside the process. It asserts that the framework exists, is voluntary, supports the June 2 executive order, and gives the federal government initial access and visibility to frontier models before wider deployment with the power to deny or restrict public access. It contains no account of what the framework requires, and so does not fill that gap.
Provenance
Captured on 2026-09-12 from the PDF hosted by Americans for Responsible Innovation, complete at three pages including the signature block and the carbon-copy list. Eight divergences and cautions against prior secondary accounts are recorded in the raw file's ## Source-fidelity notes.
Relationships
- related: Americans for Responsible Innovation (ARI) — the letter's lead signatory and host
- related: Center for Democracy & Technology (CDT) — second signatory
- related: AI Pre-Release Vetting — the confidential federal review process the letter seeks disclosure of
- related: Independent Verification Organizations (IVOs) — the question of who verifies a frontier model before wider deployment
- related: Regulatory Capture in AI Policy — the market-structure harm the letter asserts
- related: Open-Source AI / Open-Weight Models — the claim that a confidential framework disadvantages smaller and open-source labs
Sources
- Raw source:
Raw Sources/Coalition Letter re Public Release of White House AI Framework (2026-09-09).md - Letter PDF: ari.us