In Red Lion Broadcasting Co. v. FCC, 395 U.S. 367 (1969), the Supreme Court upheld the “fairness doctrine,” which required every broadcaster “to give adequate coverage to public issues” that is “fair in that it accurately reflects the opposing views.” The doctrine was enforced primarily through the Commission’s licensing authority. The Court, acknowledging that the doctrine granted the FCC extraordinary authority to regulate speech and the press, held that such power was appropriate because radio and television frequencies are inherently scarce and the government thus had an obligation to regulate the content of broadcasts to ensure fair and balanced coverage on issues of public importance.

Unbeknownst to the justices, however, the Red Lion case originated in a “politically motivated campaign to use the fairness doctrine to harass stations airing right-wing commentary, an effort inspired and managed by the White House and the Democratic National Committee and largely financed with political contributions.” The Nixon administration and its allies then used the same playbook to lean on broadcasters perceived as unfavorable to the President. The FCC abandoned the doctrine during the Reagan administration, partly in recognition of the danger such regulatory authority posed to press freedom.

Although the fairness doctrine is now defunct, it underscores how the FCC’s powers to regulate broadcast licensees can be used to suppress viewpoints perceived as unfavorable to the party in power.

Much has changed since the Court decided Red Lion. The audience for over-the-air broadcast is shrinking while internet-based media like streaming grow in both viewers and listeners as well as influence on issues of public concern. (Simultaneously, broadcast remains an essential medium for huge numbers of Americans, especially those of fewer means, making government interference in content decisions that much more fraught.)

The conditions that previously supported the Supreme Court’s decision in Red Lion have thus completely changed. The Government, particularly an agency with potent licensing and merger review authority, should have no power to regulate broadcast content beyond any that the government currently possesses for any other expressive medium (namely, the power to regulate unprotected speech or restrictions that meet strict scrutiny, which few will). Congress should clarify as much in statute, or a challenge should be brought to the Supreme Court to overrule Red Lion.

Indeed, Red Lion has never been disavowed, although the justices have expressly declined to extend it to other, later-developed communications media, including cable television (Turner Broadcasting v. FCC) and the Internet (Reno v. ACLU), to which the “scarcity” rationale for regulation is plainly inapplicable.

As long as the Red Lion precedent stands, every president will have the authority to assert control over broadcast content. This threat has only been magnified under the current administration.

In response to congressional questioning regarding whether the FCC might take enforcement actions based on constitutionally-protected editorial decisions by media companies, Chairman Brendan Carr stated that the FCC has “walked away from enforcing the public interest standard, and I don't think that's a good thing.” He also cited the news distortion policy and the broadcast hoax rule, which derive from the public interest standard, and stated that his position “is that we should be enforcing those rules and policies.” And, in January 2026, the FCC issued guidance suggesting that it would narrow the “bona fide news” exception to the equal time rule, which permits candidates whose opponents are afforded access to a licensee’s facilities to demand an equal opportunity for airtime.

This May, the FCC issued a Public Notice to “remind broadcasters of their longstanding public interest obligations and further ensure that broadcasters are continuing to comply with the public interest obligations that underpin their licenses.” The notice opened with a reiteration of the spectrum scarcity rationale that grounded the Red Lion opinion. It further stated that although the FCC has reduced many burdensome broadcast regulations, “the role of the Commission in determining whether broadcasters have met their public interest obligations has not faded.” And, emphasizing the Commission’s public interest authority, the notice asserts, “Where the Commission finds that a broadcaster has failed to serve the public interest, the Commission may take appropriate action, including enforcement action, grant of a renewal application with conditions and/or on a short term basis, requiring a licensee to file an early license renewal application, or designating an application for hearing.”

Given the danger of direct censorship inherent in the misuse of the public interest standard, Congress should exercise its responsibility to preserve the free flow of newsworthy information to the public by clarifying the scope of the standard to prevent overreach by the Commission.

Another approach could be a legal challenge seeking to overturn Red Lion. An opportunity for such a case may be on the horizon, as the same day the FCC issued its May Public Notice, ABC filed early license renewal applications for its eight owned-and-operated stations pursuant to an April 28, 2026 FCC order. ABC included with each application a detailed public interest statement and an objection that describes the FCC’s early renewal mandate as an “effort to suppress free speech under the guise of bureaucratic process.”

However the law changes, Chairman Carr’s cloaking of his efforts to influence broadcast coverage in the public interest standard make it clear that it must change. Indeed, fewer things could be more outside the public interest than an FCC interfering in what we see and hear over the air.

The following can be attributed to Scott Wilkens, senior counsel at the Knight First Amendment Institute:

“President Trump and his former co-defendants should not be allowed to delay this case further. This report concerns allegations of serious misconduct by the nation’s highest public official, and the public has an extraordinary interest in seeing it. The court should reject this latest attempt at delay and hear the case on November 2.”

The case concerns Volume II of Special Counsel Jack Smith’s report, which details his investigation into the defendants—President Trump, his personal aide Walt Nauta, and Mar-a-Lago property manager Carlos De Oliveira—over the alleged mishandling of classified documents at Mar-a-Lago. In February 2025, the Knight First Amendment Institute and American Oversight separately sought access to the report, arguing that the public has a right to see it. After months of litigation—and a ruling by the U.S. Court of Appeals for the Eleventh Circuit that there had been “undue delay” in the district court proceedings—Judge Aileen Cannon permanently blocked the report’s release in February 2026 and denied the Knight Institute’s and American Oversight’s motions to intervene. The Knight Institute and American Oversight appealed. On September 30, the Eleventh Circuit set oral argument for November 2, after the appeal had appeared on several tentative argument calendars. In papers filed today, the Knight Institute and American Oversight argue that there is no justification for delaying the argument again. 

Read today’s brief in United States v. Trump here.

Detailed information on the case, including a timeline of significant court filings, is available here.

Lawyers on the case include Scott Wilkens, Jameel Jaffer, and Alex Abdo for the Knight First Amendment Institute, and David Buckner for Buckner & Miles.

For more information, contact: Lorraine Kenny, lorraine.kenny@knightcolumbia.org. 

What all our names collectively convey is that we are organizations of journalists that exist to defend the rights of our colleagues—that is the “press” in press freedom.

I have often been intrigued by the name of the leading press freedom group in Canada, which is called Canadian Journalists for Free Expression (CJFE). Consider for a minute what that name communicates.

In the U.S., the framing and public perception is that journalists and media institutions come together when their own rights are threatened. Their mandates are grounded in a conception of press freedom necessitating endless debates about who is a journalist, where the profession begins, and where it ends.

CJFE meanwhile says it defends “free expression and those with the courage to seek and report the truth.” Its mandate is more flexible. Its historic mission is grounded in the broader defense of speech rights, encompassing Canadians who express themselves in a variety of ways.

In the U.S., the subtext is, “We as journalists have come together to defend our rights and we invite the public to support us in this endeavor, because everyone benefits from the work that journalists do and when information circulates freely.”

In Canada, the subtext is, “The rights that guarantee the free exercise of journalism are the same ones that protect speech rights exercised by all Canadians—so when your rights are threatened, our rights are threatened. We as journalists stand with you.”

Recently, I’ve begun to wonder if that broader framework might make sense in the U.S. Instead of asking the U.S. public to stand with journalists when our rights are threatened, perhaps we as journalists and media institutions should find ways to stand with the public. This approach may have particular resonance in the U.S. at a time when the media environment is evolving rapidly and the line between journalism and other forms of expressive activity are becoming ever more blurred.

Flipping the script in this way could strengthen trust and engage the public at a time of deepening polarization. The traditional journalist’s refrain—“we are your eyes and ears” or “we are your surrogate and represent the public in places to which you don’t have access”—is less likely to resonate than “we stand with you in defending your rights.”

Second, it would better reflect the way journalism is increasingly exercised. In fact, it is often members of the public who serve as the eyes and ears of the media, providing firsthand documentation that journalists then verify, contextualize, and disseminate to a wider audience. This was apparent during Operation Metro Surge in Minneapolis in January, when reporting on police abuse, including the killing of two protesters, was based largely on cell phone footage uploaded by bystanders. While journalists have a distinct professional identity and social function, the legal framework that protects their rights while covering protests is the same one that protects the protesters themselves.

Third, this broader framing would reflect the legal reality of how journalists actually defend their rights when they go to court. For the most part, journalists do not have and do not claim any special rights. When media organizations engage in litigation to defend their rights, they usually argue that the free speech principles that protect the rights to all Americans must apply to journalists as well.

Of course, the counterarguments are also compelling. Broadening the mandate in the Canadian style could undermine the media’s perceived independence and drag it into unwanted controversies. It could erode professional solidarity and weaken funding structures, since press freedom organizations in the U.S. are often backed by the media industry and industry-aligned foundations. There are myriad practical questions. What kind of free speech cases might media and journalistic organizations take up? Who would raise their voice, the news organizations themselves or the press freedom groups that they support? What kind of interventions are appropriate? Public statements? Amicus briefs? Public policy advocacy? In fact, CJFE’s own experience is an object lesson in what can go wrong. The organization has not fully recovered from its near collapse in 2018 after the staff issued a statement condemning Israeli violence against journalists and protesters in Gaza.

But times are very different today, and so is the practice of journalism. By maintaining a strictly domestic focus—standing up for Americans exercising their First Amendment Rights—journalists and the institutions that represent them can help shift the public debate in this country away from the institutional role of the media and more toward the exercise of our collective rights. In terms of issues, media organizations might consider supporting legal efforts to defend the rights of all Americans to engage in protest, including the right to record police activity. That opportunity was brought home by a recent decision in the Second U.S. Circuit Court of Appeals brought by a non-journalist and involving activity that was not fundamentally journalistic. Media organizations could also consider the possibility of joining efforts to defend the speech rights of non-citizens which are constitutionally protected and under assault.

A broader defense of speech rights by U.S. media institutions would recognize that the distinction between journalism and other forms of expressive activity are becoming more difficult to parse. It would align more closely with the legal reality, which is that journalists do not enjoy “special” rights and do not generally seek them. And it would reflect the obvious truth that, when it comes to freedom of expression, we are all in this together.

“The court’s decision is disappointing, but we intend to appeal. Spyware manufacturers that participate in the persecution of journalists shouldn’t be able to operate with impunity, and U.S. courts must ensure that they are held accountable for their actions where those actions violate U.S. law,” said Carrie DeCell, senior staff attorney and legislative advisor at the Knight First Amendment Institute. “We will continue to seek justice on behalf of El Faro journalists in U.S. court, where the case belongs.”

In November 2022, the Knight Institute filed the lawsuit alleging that NSO Group violated U.S. law by developing, selling, and assisting in the deployment of Pegasus against El Faro. Between June 2020 and November 2021, El Faro’s employees were subjected to at least 226 Pegasus infections. Through these attacks, their iPhones were accessed remotely and surreptitiously, their communications and activities were monitored, and their personal data was accessed and stolen. In recent years, NSO Group’s spyware has been used by authoritarian and rights-abusing regimes to target journalists, human rights activists, and political dissidents around the world. This case is the first filed by journalists against NSO Group in U.S. court.

Read today’s decision here. 

Read more about the case, Dada v. NSO Group, here.

In addition to DeCell, lawyers on the case include Jameel Jaffer, Alex Abdo, Stephanie Krent, and Nicola Morrow for the Knight First Amendment Institute, and Corey Stoughton for Selendy Gay PLLC.

As is true in most other areas of life in 2026, AI is being floated as the solution to FOIA’s woes. The federal government is already bullish on it—the Department of Homeland Security, the National Archives, State, and Justice, among others, have dabbled in its use. These experiments can be understood as a response to both short- and long-term trends, from a depleted federal workforce to the ever-increasing number of requests submitted across the federal government. A record 1.7 million requests were submitted last fiscal year, a million more than a decade ago. And this all comes as the number of federal records being created and stored is exploding. Even if agencies’ FOIA resources had remained constant into the current administration (which they didn’t), the government would be falling behind.

The allure of AI to address these trends is understandable. As researchers like Jason Baron have noted, the ever-increasing volume of federal records makes AI-assisted searching a particularly attractive idea. Given how fast AI is progressing, though, it’s not hard to imagine it handling the entire FOIA pipeline: interpreting requests, locating records, reviewing and redacting them for exemptions, and responding to the requester. Deployed at scale, this technology holds out the promise of reviewing millions of pages of text, understanding decades of law, and tracking factual minutiae all without complaining about being sent to Siberia. In this vision of the future, AI is a wholesale replacement of the administrative process, cutting through the trends that had seemingly doomed FOIA. Assuming such a vision is technically possible, though, is it something we want to work towards?

Something clearly needs to change. After more than a decade of litigating requests for journalists and news outlets, I am acutely aware of FOIA’s problems and their effect on newsgathering. And after talking with countless government employees at mediations, conferences, and committees over the years, I understand many of the complexities within the government that have contributed to FOIA’s delays. But before introducing a radical new technology we must be clear-eyed about what we are trying to achieve, how we will mitigate risk, and the red lines that cannot be crossed. In short, we need a FOIA AI framework. I have endeavored to sketch one out here.

My proposal has two parts:

• First, we must assess the risk of introducing AI into any particular part of FOIA. The European Union’s AI Act provides a good basic formulation: what is “the probability of an occurrence of harm and the severity of that harm”? Risks should be categorized, and the most severe ones mitigated.
• Second, we must empirically test whether the core values of FOIA are being promoted by a particular implementation of AI. In a nutshell, those values are disclosing as much non-exempt information to the public as quickly as possible. We could test, for example, whether introducing an AI-powered search tool within an agency actually leads to faster request response times.

I want to focus on the first part of the framework in this piece, because the line from this White House has generally been an all-gas, no-brakes approach. We are charging into widespread AI adoption even though, as described below, there is serious risk to FOIA and not much theorizing about the nexus. In 2024, a subcommittee of the FOIA Advisory Committee (on which I served) called for the Department of Justice to issue guidance to agencies on their use of AI in the FOIA process. To date, that has not happened. And the Office of Management and Budget’s 2025 memo on AI, which calls for additional steps when it comes to “high-impact AI” within the government, does not include FOIA processing among its presumptive categories. That is a mistake that should be rectified.

A core part of risk management is evaluating the severity of harm that might result from any particular course of action. When it comes to FOIA, we should break out different parts of the administrative process for evaluation: on one end of the spectrum there are many basic, repetitive tasks where the potential harm is low. This includes actions like writing to the requester acknowledging receipt of their request, responding to basic queries from the requester (where is the request in the process, what is the estimated date of completion), and document management (like sorting and de-duplication). I classify these as low risk both because AI is pretty good at them and because the consequences of getting it wrong are not very serious.

At the other end of the spectrum are decisions at the core of FOIA: how to search for records, what to release, and what to withhold. The worst-case “harm” of getting it wrong here is the requester failing to get what they asked for. That could be disastrous for persons in immigration proceedings, who must generally FOIA their own files from the government. For the press and civil society, it’s going to be the deprivation of records promoting accountability and informed self-governance. These scenarios are unquestionably “high-impact”: FOIA has helped expose COINTELPRO, the torture of prisoners by the US government after September 11, the details of the Federal Reserve’s secret liquidity program, the Afghanistan Papers, the toll of civilian casualties from US-led airstrikes in the Middle East, the Robert Mueller investigation records, and so many more stories. These records have had profound implications in both domestic and foreign affairs, and that type of accountability must be preserved.

What is the likelihood of these most severe harms if AI were deployed to answer FOIA requests? So much depends on the technical details of the deployment, which are impossible to predict ahead of time. There are also endemic problems with AI, like its penchant for hallucinations, and those could certainly create problems in processing a request. But it is not hallucinations that worry me the most. It is the fact that AI can appear to be neutral and objective while it actually obscures and magnifies prejudices to a greater extent than other technologies. Each model contains biases from its training corpus, its fine-tuning, its background instructions, its prompts, and so on. Those reflect human biases stemming from deliberate decisions, predispositions, and prejudices. And they will, in turn, be imported into whatever system AI is inserted into.

Consider the State Department’s machine-learning declassification trial program. The system was not trained to produce an objectively “correct” result, but rather to match the decisions of the agency’s human reviewers. The same is true of an AI tool Health and Human Services has deployed, called “FRED,” that purports to “analyze, identify, and generate predictions of text for redaction” under FOIA. FRED was trained on certain inspection “forms before redaction and versions of those forms after redaction by FDA staff,” just as at State. The government might argue that AI matching the results of the human review is the correct result, but we know that FOIA officers routinely overredact records when reviewing them for release. If AI bots are trained against the output of human reviewers, similar over-redaction tendencies will emerge.

We should also consider that more insidious directions could be built into these systems. The current administration, for example, instructed FBI agents to “flag” any documents in the Epstein files that mentioned the president. Secretary Clinton set up a private email server, establishing a whole category of records that was never searched. ICE recently put out a purported policy of only releasing bodycam video when it is in the agency’s “best interests.” And so on. A bot that is trained to redact politically inconvenient information at scale, or instructed to never search a particular record repository, would pose grave risks to FOIA’s transparency mandate.

If FOIA is already experiencing these types of issues, though, perhaps a fairer measurement of risk is the delta between a human-centric FOIA system and one powered by AI. Even here, my sense is that it will be significant. Right now, line-level FOIA officers make the initial decisions about how to search for records and what to redact. Those decisions are subject to review and sometimes are overruled by higher-ups, to be sure. But in most cases the basic contours of disclosure are set by persons who have some distance from the White House, are trained to follow the law, and can push back against political meddling. A human reviewer might balk at a blatantly political instruction or blow the whistle about it, as happened with the Epstein files. Or humans might simply release the information anyway, as the Office of the Director of National Intelligence did last year with a memo on Venezuela.

Aggressively deploying AI could invert this structure, giving the White House near-complete control in FOIA responses from the get-go. If a FOIA bot is deployed to make the first cut on redactions, it could be given instructions like: redact anything that contradicts the president’s statements. AI can be trained from the outset to accept these types of instructions, and even if a human reviewer is “in the loop” somewhere they might never know that political instructions were built in. AI is so good at coming up with plausible-sounding explanations that a FOIA bot tuned to a particular set of goals could generate dozens of pages purporting to justify any redaction, even if it is contrary to black-letter law. Deployed at scale, it could cloak instructions by burying them in output.

Obscuring these types of instructions goes hand in hand with another major difference AI would bring: “automation bias.” This is the tendency of humans to “defer to automated systems,” leading to a reduction in “the amount of independent scrutiny that they exhibit when making decisions.” This is a well-documented phenomenon because it turns out that “people are bad at judging the quality of algorithmic outputs and determining whether and how to override those outputs.” If AI were to displace some or most of the FOIA processing pipeline, even humans somewhere in the loop might simply defer to whatever the AI says. Studies are already starting to show that using AI can reduce critical thinking, a finding that recalls Lisanne Bainbridge’s observations about the challenges of overseeing mechanized industrial processes in Ironies of Automation. We should be concerned that such tendencies, over time, will consolidate the locus of FOIA control higher up within the government.

Mitigating these risks will require extending current structural protections in FOIA. Congress wasn’t blind to the fox guarding the henhouse problem in creating this law, so review of the agency’s initial decision is already baked in, first through an administrative appeal and then de novo judicial review. The meaningfulness of these procedures must be preserved, both by disclosing additional information about AI’s usage and by limiting when it can be deployed.

My first set of risk mitigation proposals concerns the executive branch. Any AI FOIA tool development should be conducted in the open and, ideally, in collaboration with non-governmental experts and representatives of the requester community. Agencies should also be required to conduct periodic audits of any AI FOIA system and publish the results with enough detail for meaningful external review. These measures should not be controversial—the Justice Department’s FOIA handbook is public, as is its written guidance to agencies. OMB already requires “high-impact” AI applications to be subject to a number of additional procedures for testing and feedback. The government should extend these to FOIA systems to mitigate systemic risks.

When it comes to processing individual requests, agencies need to go further. Specifically, they must:

• Inform requesters whenever AI is used to respond to their request;
• Inform requesters how AI was used to respond to their request (e.g., for searching, for de-duplication, for exemption application);
• If AI was used to locate records, inform requesters as to which government databases/record repositories were searched;
• Preserve all of the inputs and outputs from AI systems that pertain to requests—the AI “administrative record”—so they can be reviewed; and
• Provide, upon request, a manual review and renewed decision regarding any action of an AI system.

These steps will not eliminate the risks of AI in the FOIA process, but hopefully they will lay the groundwork for ensuring it does not have the last say.

The second set of proposals concerns litigation. Judicial review is far from perfect (judges tend to agree with agencies most of the time), but it is the backstop for all of FOIA’s accountability. Indeed, the most impactful records released over the years have tended to come out in litigation. Preserving that independent review is a red line, and where AI’s involvement within the government must end. A situation where an agency can only gesture at the AI and claim it’s tokens all the way down cannot form the factual basis for judicial review. Once a case commences, defendant-agencies must conduct a manual review of their decisions and limit their litigation posture to the results of that human review.

What AI can do to improve the FOIA process is a question that should be discussed, experimented with, and empirically measured. But we must have guardrails in place to protect the law’s purpose first. Properly assessing and mitigating risk in the FOIA pipeline is overdue.

“I’m delighted to welcome Vanita Gupta and Karen Kaiser to the board as the Knight Institute enters its second decade,” said Lee C. Bollinger, president emeritus (ex officio) of Columbia University and chair of the Knight Institute’s board of directors. “Over the past ten years, the Institute has made important contributions to the defense of freedom of speech and the press. Those freedoms now face renewed and significant challenges, and Karen and Vanita will strengthen the board at a critical moment for our work.” 

The Knight Institute was established by Columbia University and the John S. and James L. Knight Foundation in 2016 to defend the freedoms of speech and the press in the digital age through strategic litigation, research, policy advocacy and public education. The Institute promotes a system of free expression that is open and inclusive, broadens and elevates public discourse, and that fosters creativity, accountability, and effective self-government.

“Karen Kaiser has spent her career protecting the people who report the news, and Vanita Gupta has spent hers making sure the law protects everyone equally,” said Jameel Jaffer, the Knight Institute’s executive director. “They are talented lawyers and inspiring leaders, and I’m delighted that the Institute will have the benefit of their experience and judgment.” 

Vanita Gupta is director of the Center for Law and Public Trust and a distinguished scholar in residence at NYU School of Law. From 2021 to 2024, she served as U.S. associate attorney general, becoming the first civil rights lawyer to hold one of the Justice Department’s top three leadership positions. Gupta previously served as president and CEO of the Leadership Conference on Civil and Human Rights and led the Justice Department’s Civil Rights Division during the Obama administration. Earlier, she was deputy legal director and director of the Center for Justice at the ACLU and an attorney at the NAACP Legal Defense & Educational Fund, where she led the effort to overturn the wrongful drug convictions of 38 people in Tulia, Texas. 

Karen Kaiser is director and legal advisor at the Clooney Foundation for Justice, where she works on harnessing AI to expand access to justice for women and journalists at risk. She previously spent 12 years as senior vice president, general counsel, and chief compliance officer at the Associated Press, where she oversaw the news organization’s global legal operations and worked on issues including press freedom, journalist safety, and AI governance. Kaiser was instrumental in reforms to U.S. Department of Justice guidelines governing subpoenas of members of the press in 2014 and 2022 and served on the DOJ News Media Task Force for more than a decade. She is chair of the International Bar Association Media Law Committee and serves on the steering committee of the Reporters Committee for Freedom of the Press. 

In addition to Bollinger, Gupta, and Kaiser, the Institute’s board includes Maribel Pérez Wadsworth, Knight Foundation President & CEO (ex officio); Daniel Abebe, Dean and Lucy G. Moses Professor of Law at Columbia Law School; Jelani Cobb, Dean and Henry R. Luce Professor of Journalism, Columbia Journalism School; Nicholas Lemann, Dean Emeritus at Columbia Graduate School of Journalism; Gerald Rosberg, former Senior Executive Vice President at Columbia; and Deb Roy, Professor of Media Arts and Sciences at MIT. 

For more information, contact: Lorraine Kenny, lorraine.kenny@knightcolumbia.org. 

Providing access and support for transparency is essential to civic engagement, self-governance, and protections against government corruption. It is the primary avenue for citizens to see what their government is up to and whether it is working in the public’s interest. However, whether by design or consequence, the federal FOIA process is nearly stagnant.

What we know:

• The number of FOIA requests is growing at an exponential rate.
• Federal FOIA staffing was heavily reduced at the hand of DOGE, and those unfilled positions mean fewer requests can be processed.
• Backlogs continue to be on the rise, as they have been for years, but they are now at their highest levels.
• FOIA professionals do not have adequate resources, specifically in staffing and technology, to resolve the backlog of requests at the federal level.
• Technology-assisted requests are making it easier for requesters to file extensive requests faster than humanly possible.
• Even if FOIA professionals want to make internal changes to increase efficiencies, they face obstacles within their agencies, slowing and blocking changes.

Access professionals are the front-line stewards of public access to government information. If they don’t have the time and tools to get their work done, information will not be publicly available. They are responsible for facilitating public access by receiving requests, tracking them, interpreting what is being requested, determining what will be released, levying fees when allowable, and resolving disputes with stakeholders. They must also ensure legal compliance through determining redactions, exemptions, denials, timelines, records management, and potential litigation. They do all of this while managing internal processing workflows—conducting records searches, reviewing and preparing records for release, keeping records organized and searchable, and following retention schedules.

It is a balance of legalities and logistics. Like any engine, it can’t run well without fuel, and in this context, that fuel is money, manpower, and technology. Agencies across the government are disparate in how they functionally respond to requests. Some are centralized, and some are decentralized. Some use advanced technology for search, review, redaction, and analytics; while others rely on spreadsheets and electronic filing. The needs of larger agencies that receive more than 100,000 requests per year would not compare to small agencies that receive fewer than 50 requests per year. But across most agencies, the backlogs are becoming too long to clear, and access professionals are just watching new requests join the queue with no hope of response any time soon.

A solution will require a multi-faceted approach to augment resources, both within agencies and through external funding structures: increase funding, analyze internal support operations, and review efficient technology use.

Increase funding

Getting the proper resources to drive more efficient processing takes money, and, at the federal level, that requires Congress to require agencies to properly fund the FOIA system. The FOIA Advisory Committee recently recommended that Congress provide funding directions that require agencies to consider their workforce size, the changes in requests in the past five years in volume and complexity, backlogs, and other operational structures (Recommendation 2026-03). In 2021, the Office of Government Information Services recommended that Congress ask the Government Accountability Office to study whether agencies are adequately funded and what they could accomplish with additional funding, specifically timely response and backlog reduction.

However, in this current administrative climate, it is unlikely that Congress as a whole will respond in a fashion supportive to FOIA. But individual lawmakers can make such requests of the GAO, or the comptroller general can initiate such studies. A potential catalyst would be (like the intent of FOIA itself) to shine light on the financial challenges agencies are experiencing.

Analyze operations

Agencies need to realistically analyze staffing levels and workloads to understand why the backlogs are increasing and what would be required to eliminate them. This analysis should also consider revising operational procedures. For example, do agencies need dedicated FOI professionals to respond to requests? Do they need centralized systems to respond? Do they need additional subject matter experts on staff to process complex requests?

In focus group research, FOIA professionals said that their work is rarely considered mission critical, and therefore they have trouble getting the proper resources they need to complete their work. As a result, the FOIA Advisory Committee recommended that agencies study the infrastructure needed to carry out their open government responsibilities (Recommendation 2026-11). Healthy FOIA operations run on proper staffing levels as well as appropriate technology to support productivity.

Review technology

Agencies should also review how they can adopt new technology to use for more efficient processing. Externally, the Office of Information Policy, the Office of Governmental Information Services, and the Department of the Interior are developing a FOIA technology inventory to evaluate what is currently in use. However, agencies themselves should study how technology can augment their own internal processes and how those processes can be modernized.

Adopting technology in government agencies is a long and protracted process. It is so difficult that technology can be outdated by the time agencies actually receive it. Technology adoption and acquisition is a slow and protracted process at the government level, and there have been previous calls for a revision of the process, especially for agency response to information requests. Additionally, there is no consistent system across agencies because agencies adopt the technology that makes the most sense for their needs and that fits their own budgets. This lack of interoperable technology across agencies is not a new obstacle, but it is becoming more apparent as other FOIA resources are shrinking.

Agencies’ piecemeal responses to the inability to respond to requests are not a long-term solution. For example, providing time extensions, a common response to the lack of capability to respond to requests, doesn’t solve the problem and disproportionately harms requesters. Agencies are simply moving the problem later in time rather than managing what is causing the problem. The requests are not the problem, the response capability is.

Charging more fees for voluminous or complex requests also ignores the root causes and unduly punishes requesters, especially when they are used as a cudgel to keep requesters at bay. Taxpayers already foot the bill for the FOIA process. Additional fees will not benefit the agencies charging them. Where fees are employed, they should be specifically structured to charge only what is necessary, leaving less room for weaponization.

Traditional solutions have not worked, and a concerted effort needs to be engaged. Fueling the response process with proper resources, staffing, and technology, all of which will need financial support, is essential. Unfortunately, that kind of tangible support is receding, and it is showing in backlogs and access professionals’ workloads. Ultimately, this failure will result in a lack of civic engagement and governmental accountability, which inevitably leads to eroded public trust and a weakened democracy.

Those legacies—and what it will take to confront them—were the focus of “The Forever War and the First Amendment,” a two-day symposium held at the Bollinger Forum on Columbia’s Manhattanville campus earlier this month that brought together journalists, scholars, lawyers, and activists to examine how the “War on Terror” transformed government power, the courts, the press, and democratic culture.

Co-hosted by Columbia’s Knight First Amendment Institute and the Simon and June Li Center for Global Journalism at Columbia Journalism School, in partnership with Harper’s Magazine and the Columbia Daily Spectator, the symposium examined government powers, legal doctrines, surveillance systems, and media practices that were in place before 9/11 and expanded afterward.

Opening the symposium, Jameel Jaffer, Knight Institute executive director, recalled the “apocalyptic feel” of New York City in the weeks after 9/11: smoke and ash in the air, fighter jets flying low over the city, soldiers in the streets, missing-person posters on every lamppost, and everyone on edge about the possibility of another attack.

“Many of us also had a sense of foreboding about how our government would respond,” Jaffer said. “Even in that moment of intense fear and sadness and confusion, it was clear that collectively we faced a momentous choice and that what our attackers did next was probably going to be less important, less consequential than what we ourselves did.”

What came next—the decisions, policies, and normalized ways of thinking that took hold in the years after—was the focus of the symposium.  

The Deeper Roots of the Forever War 

In her remarks, journalist Rozina Ali noted that she had expected her recently published book on Islamophobia, Seasons of Fury, would begin with 9/11, but her research led her to surveillance and law-enforcement practices going back to the 1960s. Artist and filmmaker Assia Boundaoui uncovered another example of pre-9/11 surveillance while making the award-winning documentary The Feeling of Being Watched and developing Inverse Surveillance Project: tens of thousands of pages of FBI files revealing that her Chicago-area Muslim community had been the target of a sweeping pre-9/11 counterterrorism investigation code-named “Operation Vulgar Betrayal.” 

“The data that is collected in the surveillance remains in perpetuity,” Boundaoui said. “There’s a double violence in this because it can be reused.” 

Across the symposium, speakers returned to a common pattern: Legal authorities and institutional practices that often predated 9/11 were expanded after the attacks and then normalized. They traced that development through the material-support laws, national-security surveillance, secrecy surrounding torture and targeted killing, the use of immigration powers against political dissent, and the weakening of institutions intended to hold executive power accountable.

When the Courts Deferred—or Did Not Decide 

ACLU attorney Brett Max Kaufman traced judicial deference to the climate immediately following the attacks: “Even judges that on 9/10 might have been inclined to be skeptical of government power or push back on certain kinds of arguments—on 9/12, that was no longer the case. That fear, the culture of censoriousness, and not knowing when the next attack was going to come affected them as well.” 

A day-two panel devoted to the First Amendment examined the consequences of that judicial deference. Panelists discussed cases ranging from Holder v. Humanitarian Law Project, in which the U.S. Supreme Court ruled that providing human rights training to a designated terrorist group could be a crime, to unresolved challenges involving warrantless surveillance and the punishment of noncitizens for political expression. Knight Institute litigation director Alex Abdo asked whether the courts had created “a jurisprudence of ignorance,” in which rights become “effectively unenforceable” because an array of doctrines allows courts to avoid deciding on the merits. 

Reporters for the Columbia Daily Spectator interviewed people whose stories intersected in important ways with the symposium’s themes. Alberto Mora, who served as general counsel of the U.S. Navy from 2001 to 2006, discussed his opposition to torture and the imperative to protect human dignity and the rule of law during a crisis. Mohamedou Ould Slahi, who was imprisoned without charge at Guantánamo Bay for 14 years and whose story was dramatized in the feature film The Mauritanian, described what happens when extraordinary powers created for supposedly dangerous people are turned against the innocent. 

How the Press Reinforced—and Exposed—the Forever War 

A panel moderated by Jelani Cobb, dean of Columbia’s Journalism School, examined how the Forever War both tested the press and exposed its weaknesses. Government surveillance, efforts to identify confidential sources, and secrecy surrounding the legal basis for torture made reporting more difficult. At the same time, news coverage often amplified official claims and helped legitimize government policies. Yet investigative journalism proved indispensable to exposing what officials had concealed. 

Dana Priest, whose exposure of the CIA black sites and other aspects of the counterterrorism campaign in The Washington Post earned her a Pulitzer, acknowledged that reporters and editors at both The Post and The New York Times failed to question official claims aggressively enough, while skeptical reporting was often given less prominence. 

A second-day panel examined how nonprofit newsrooms including Airwars, Drop Site News, and Zeteo are developing alternatives to legacy coverage, often through collaboration with local journalists and open-source investigators. Haley Willis, of Columbia Journalism School, described open-source reporting as “taking tools that kind of started in the hands of governments and militaries and turning them on them,” while cautioning that access to satellite imagery and other evidence can still be limited by cost and corporate or government pressure. 

The first day closed with a moving 45-minute performance by renowned pianist and composer Vijay Iyer, featuring work shaped by his response to 9/11 and its aftermath. Introducing the performance, Iyer described music as a way of reaching places words cannot, calling what followed “a kind of séance.” 

In connection with the symposium, the Knight Institute and Harper’s Magazine co-commissioned an essay by the writer Hari Kunzru about art created during this period. At the symposium, Kunzru joined journalists Steve Coll, M. Gessen, Nicole Hemmer, and Phil Klay for a panel about “The Forever War and Democratic Culture.” Kunzru’s essay in Harper’s is available online. 

Readers can watch recordings of individual panels of the Forever War symposium on the Knight Institute’s YouTube channel.

With the start of a new academic year upon us, as approximately 150,000 students are in their first weeks of law school, I’m writing a series of reflections on the conversations we had with podcast guests, thinking about the future of legal education and of professional responsibility.

For our first episode, Madhav and I talked with Kim Lane Scheppele, a leading expert on illiberal autocracy and democracies under stress. She described how the election of populist leaders, like Viktor Orbán in Hungary, is often followed by efforts to use the law to undermine constitutional institutions, and why this can draw lawyers into an authoritarian project even if they don’t intend to support it:

When you put into the process laws that actually have the effect of undermining democratic institutions, you get a lot more buy-in than if you simply did it by edict or by force, and so it legitimates the entire enterprise of undermining democratic institutions. That’s why law becomes so crucial. Now, who are the lawyers who do this stuff? Some of the lawyers who do this are lawyers who believe in the leader and believe in the program. But often times you get lawyers doing this because law is like a giant math problem if you like math.

It’s just fun to play with the rules and think, well gee, you know, how would you do this? How would you do that? Is there a way you can make an argument for this, that, and the other thing, and frankly, legal education in many places, and here I include the U.S., lends itself toward exactly this kind of lawyering. Because how often do we take first year law students and say, argue the opposite of what you believe…. So lawyers have learned to be hired guns. They’ve learned to think that clever legal arguments are good lawyering, and they have much less training and much less grounding in the values that underwrite democratic institutions, so that the lawyers themselves may not realize how much they're undermining democratic institutions with the very things they're recommending. (11:17-12:16, 13:07-13:32)

I’ve continued to think about Kim’s observation that many lawyers treat legal argument like they’re solving a math problem. There is nothing new, or inherently malicious, about lawyers making clever arguments in favor of their clients. Much of legal education is devoted to learning how to do this. Indeed, lawyers have an ethical duty to “zealously represent” their clients. However, lawyers are generally not trained to consider the potential consequences for democracy of their legal arguments—to ask themselves whether the legal rule they are advocating for may undermine the foundations of our democratic system. This is where at least some of our problems arise. And this makes me recall something else we discussed during our conversation with Kim: Many lawyers in the United States simply can’t believe our democracy could fail, that it is breakable.

This ties into the myth of American exceptionalism, of course. Most U.S. lawyers are educated in a system that inculcates the idea that we have a uniquely strong democracy, that our constitution is the best of all possible constitutions, and by extension, that though our system of government may be sorely tested, it will never truly come apart at the seams. If one believes that our democratic system will endure forever, come what may, then we’re all just playing a game to be the cleverest lawyer in the room.

Instilling in lawyers a sense of the ultimate fragility of democracy, and of their individual duty to protect it, is a project that will require the participation of many institutions, but law schools must play a fundamental role in it. Law students hear plenty about the constitution and at least a little about their duty to defend it. More rarely are they asked to seriously consider what duties they, personally, bear to uphold our democratic system, and most law students are given very little instruction about the conditions under which democracies erode or fail. I wonder if constitutional law classes would land differently if students learned that the United States only met the basic requirements for a democracy after the Voting Rights Act was passed in 1965. Or if every 1L took a capsule course on how Hungarian democracy was dismantled through legal maneuvering under Orbán. I hope you’ll listen to the full episode for more of Kim’s insights on the role of legal education in shaping lawyers who will stand up for democracy, and lessons from democratic backsliding in Europe.

The following can be attributed to Jameel Jaffer, executive director, Knight First Amendment Institute at Columbia University:

“We welcome Judge Kelly’s order. Trump’s ban harms both the press and the public, which relies on journalists to report independently on the president and his administration. We hope that Judge Kelly will soon make his temporary order permanent, and that his permanent order addresses the news organizations’ First Amendment claims, not just their Fifth Amendment ones. The First Amendment bars the president from transforming the press corps into a propaganda machine.”

The Knight Institute and RCFP’s amicus brief argues that the press has, for over a century, fulfilled a crucial role by reporting on newsworthy events from within the White House on behalf of the American people, and that the president’s decision to ban news organizations from the White House based on the viewpoints expressed in their reporting violates the First Amendment. Read the brief here.

Read today’s temporary restraining order here.

Read more about the case, CNN v. Trump, here. 

The Knight Institute has supported the Associated Press in its challenge to its exclusion from the White House, filing three separate amicus briefs over a period of 18 months. The Institute has also supported The New York Times in its challenge to restrictions on access to the Pentagon. Earlier, the Institute successfully sued President Trump over his blocking of critics from his Twitter account, which the Institute argued was a “public forum” for First Amendment purposes.

For more information, contact: Lorraine Kenny, lorraine.kenny@knightcolumbia.org. 

Of course, we need both stances. It’s hard to build something new when you’re getting run over and run off the field. Defense alone can’t produce a truly independent and in-depth press, free from undue corporate and government influence. But if we stop defending against blatant censorship and runaway consolidation, there’s little room left for viable alternative outlets.

The U.S. populace may not hold the media in particularly high regard at the moment. We face a fractured and truth-challenged world, with Trump poisoning public discourse and launching a decade’s worth of attacks on the press as “an enemy of the American people.” Yet, even as this authoritarian president attacks specific journalists, he retains the fig leaf of providing his followers with the “truth” on social media and decries only what he labels “fake news.”

So, if people agree that they want reliable facts and news—even if they can’t agree on what counts as a fact anymore—why do we rely on failing, advertising-driven business models and continually under-invest in the production of quality journalism and civic information? What problems stand in the way of producing this public good?

Trump himself, and his entire administration, are obviously a searing threat to journalism and to individual journalists’ livelihoods, safety, and freedom. Threats to shut out, silence, assault, or jail journalists are hardly a new or uniquely American phenomenon, but the rapid crescendo of such attacks under the second Trump administration is beyond alarming. At Free Press, we work to educate people about such harms, pass policies that protect against them (like shield laws and jawboning protections), and defend individual journalists from arrest and attack.

Free Press also fights corporate media consolidation, even though we don’t believe corporate media alone will produce journalism that serves all of us. There’s a tight nexus between authoritarianism and corporate consolidation. Merger parties will capitulate quickly to a corrupt regime in order to get their transactions approved. (It’d be nice to think that the fourth estate’s ethics make it immune to such tactics, but in an era of the Ellison family promising favorable coverage of the administration in exchange for antitrust approvals, we can’t fool ourselves.)

Recent developments like Disney standing up against the FCC’s threats to its ABC broadcast licenses, and even going on offense about the chilling impact of that agency’s actions, show that the tide may be turning. Big companies may have learned to stop routinely retreating, capitulating, and complying in advance when this administration pushes too far.

Yet even if the largest and most well-heeled outlets can stand up to those threats, there are still reasons to be concerned about media consolidation in authoritarian times.

For one, having fewer competing viewpoints and competing journalists is worse for expanding our horizons and uncovering the truth. For two, having fewer strong voices in the marketplace for ideas gives a government bent on censorship and narrative control a smaller number of chokepoints it needs to squeeze and coerce.

What’s more, even without an image-obsessed autocratic president overtly demanding more favorable coverage of himself and his administration, corporate consolidation is bad news. At Free Press, we’ve documented when fighting mergers and regulatory rollbacks that antitrust laws and broadcast ownership limits are necessary and effective in preserving local journalism, competition, and diversity.

The promise that larger conglomerates will take their excess profits and pour them back into improved and increased coverage is a phony, trickle-down economics argument. Increased hours of news after a television merger often means nothing more than duplicate stories and repeat newscasts, choked with watered-down national stories instead of truly local reporting. Companies insulating themselves from competitive pressures through mergers and acquisitions of rivals won’t invest more, they’ll just use the profits to pay shareholders or pay down the massive debts they ran up to make the merger happen in the first place.

Broadcast journalism and media is special, insofar as rules limiting ownership, reach, and even content have generally survived constitutional scrutiny till now. The “public interest standard,” based on a spectrum scarcity rationale, has meant that the FCC can to some degree decide who speaks over the nation’s airwaves. Brendan Carr has become Trump’s censorship czar, and whether cynically or fervently or a toxic mixture of both, he’s bent that standard perhaps beyond the breaking point in less than two years of Trump’s second term.

It’s even more difficult, in American jurisprudence and under the First Amendment, to do anything about the size and reach of corporate media companies that don’t use broadcast licenses. Free Press has used antitrust tools to beat media and tech mergers, including but not limited to broadcast deals. A storied line of U.S. Supreme Court cases confirms that undue economic concentration can violate federal antitrust law, even if the commercial “product” is protected by freedom of speech and freedom of the press. And there have been antitrust consent decrees and limits on non-broadcast industry sub-sectors, like film studio control of movie theaters and television network control of TV show production. But calls to strengthen or change our antitrust laws, to account not only for consumer welfare and workers’ rights but also political economy and even news concentration, haven’t resulted in any statutory changes yet. And those changes would be tested politically and constitutionally if and when they’re proposed.

In other words, advocates have tools to prevent undue concentration, but there’s no way to prevent all mergers and set a strict limit on reach. And even if there were, many First Amendment advocates would argue strenuously against such limits, as a cure that is worse than the disease of corporate consolidation—or at least dangerous in different ways.

Because we can’t limit the size and reach of all corporate journalism companies, and we recognize the pitfalls and dangers in trying to do so, Free Press has never been solely focused on striving for a more perfect corporate media. We need a more complete view of the journalistic ecosystem to compensate for the limits of corporate owners, publishers, and speakers.

We’ve written about how journalism is a public good that will be under-produced if it depends solely on an attention economy and sensationalist appeal, especially when the number of online content choices from human creators (and now AI too) is astonishingly large. That puts pressure on traditional cross-subsidies for news by advertising-supported television and print companies. And that prompts harmful calls to move money away from extractive tech platforms right back to the hedge funds and corporate conglomerates running the country’s largest newspapers and TV station groups. Free Press Action opposes these kinds of bills at the federal and state level too.

It’s also readily apparent that “if it bleeds, it leads” local news was never healthiest for the local communities doing the bleeding, and that a focus on the crime beat and other fear-mongering to get eyeballs on the nightly newscast perpetuates racism but doesn’t produce the kind of civic information people need.

That’s why Free Press has always seen public media as a key complement to corporate outlets. Communities benefit from having a wide range of community-centered, non-commercial, and non-profit newsrooms, with at least the potential for government funding. Public money is not anathema to a free press, so long as there’s a firewall between the legislative decisions on such funding and the editorial decisions made by its recipients. An ecosystem that includes robust public media improves the quality and amount of accountability journalism produced.

It’s no accident that an authoritarian Trump administration finished the job started decades ago by Republican lawmakers to zero out federal funding for public broadcasting. Free Press fought that rescission. More than PBS, NPR, and the largest non-commercial stations, those cuts hurt stations serving less densely populated, less privileged, and less affluent communities, as well as independent filmmakers and documentarians whose work finds a home on public media. Yet the prior Corporation for Public Broadcasting was never enough. We need to reimagine public media, to make it more decentralized, truly local, and resilient to political funding decisions.

That doesn’t mean taking government funding out of the equation, and forcing non-commercial stations to rely solely on “underwriting” or charity from well-to-do donors. It means projects like the New Jersey Civic Info Consortium that Free Press helped conceive and create. These kinds of models are gaining momentum in other states too. The idea is to gather public money from general treasury revenues or dedicated tax and funding streams, then administer newsroom grants independently to foster the production of civic information and responsive journalism.

There’s no silver bullet or single answer to preserving quality journalism against the dangers of corporate consolidation and government censorship. But public dollars drawn from diverse sources and administered independently can fund non-commercial journalism by and for the people in communities never fully served by corporate media outlets.

One of the most common refrains about the federal Freedom of Information Act (“FOIA”) within the requesting community is that the statute feels broken. Even before the changes that have occurred under the current administration, agency FOIA programs were understaffed and underfunded. There has been a trend toward less disclosure and greater secrecy that transcends party lines. And apart from abuse of exemptions, agencies have long sought to erect extra-statutory procedural obstacles like sensitive review and equities-based consultations, or creative schemes such as “still interested” letters and administrative closure.

Congress has not been oblivious to these developments. The FOIA Improvement Act of 2016, for example, was designed to ameliorate many of the problems requesters face today. The Act codified a foreseeable-harm standard and twenty-five year “sunset” rule for the deliberative-process privilege. It introduced a “rule of three” to require affirmative disclosure of frequently requested records in online “reading rooms.” Fee limitations based on the failure to follow the statute’s deadlines were intended to incentivize faster processing. Yet, despite the merit of those changes, they have proven inadequate. A decade later, in the face of a record-breaking number of incoming requests, FOIA hardly seems fixed.

For some requesters, litigation is the only way out of this mess. Congress tasked the courts with enforcing agencies’ disclosure obligations. But, in practice, judicial review falls short of Congress’s design. As I have explained elsewhere, in a proposal for a dedicated Article III “FOIA Court,” “the volume of FOIA litigation nationwide has outpaced the ability of federal courts to deal with those cases in a timely manner.” FOIA suits often fail to receive the attention they deserve. More troublingly, a certain functionalism seems to prevail among jurists. Judges often resolve cases based primarily on uncritical review of agency declarations, with limited adversarial testing. Judicial review is plagued by policy concerns, information gaps, and doctrinal ambiguity.

We need to restore the independent judicial role that FOIA promises to the public. FOIA’s shortcomings are not the result of formal Chevron-style deference. The statute expressly prescribes de novo review and, for the most part, judges review legal questions in that manner. FOIA’s judicial-review problem instead arises from a functional deference driven by jurisdictional and remedial uncertainty, the anomalous procedural architecture of FOIA litigation, and atextual judicial doctrines. Congress needs to address these failures together, through both corrective clarifications of the statutory text and substantive reforms that improve the litigation process.

Rigorous judicial review would not only provide a robust check on agencies but also offer relief to aggrieved requesters. It would positively influence FOIA processing more broadly, too. Although only a small number of requests end up in court, caselaw developments and litigation practices heavily influence FOIA processing at the administrative level. Everyone benefits when courts are doing their job.

Clarifying Subject-Matter Jurisdiction

As the Supreme Court has explained, “without jurisdiction a court cannot proceed at all in any cause.” FOIA, for instance, provides that a federal court “has jurisdiction to enjoin [an] agency from withholding agency records.” Frustratingly, some agencies argue this language implicates the court’s authority to hear a case in the first instance. But that interpretation improperly transforms the elements of a FOIA claim—an allegation “an agency has (1) ‘improperly’; (2) ‘withheld’; (3) ‘agency records’”—into a set of jurisdictional prerequisites.

This confused position, if widely adopted, would have serious implications. At the least, it requires a requester to prove the merits of a claim at the pleading stage, thus reversing the burden placed on an agency to sustain its action. That is problematic for two reasons. First, because it contradicts the well-established principle that “[t]he burden is on the agency to demonstrate, not the requester to disprove,” that the records at issue are not subject to FOIA or have not been improperly withheld. And second, because most requesters, as a practical matter, have no access to the facts needed to prove or disprove the elements of either their claim or the agency’s affirmative defenses. The government holds all the cards. (We will return to this problem of “informational asymmetry” in a moment.)

Although three circuits have rejected this jurisdictional argument, the Department of Justice continues to advance it to the detriment of requesters. Providing clarity on jurisdiction—and putting an end to the government’s game of shifting evidentiary burdens to requesters at the complaint stage—is essential for making sure FOIA litigation functions as Congress intended.

Explicating Remedial Authority

A right of access without an effective remedy for noncompliance is meaningless. Although FOIA authorizes the courts to “order . . . production” and “enjoin . . . withholding[s],” contemporary cases frequently raise broader remedial questions, which are construed as constructive withholdings to fit within this textual scheme. This is particularly true with the allegations underlying so-called policy-or-practice claims, which depend on the notion that “FOIA imposes no limits on courts’ equitable powers in enforcing its terms.” It is also true with claims involving FOIA’s affirmative-disclosure requirements.

As to the latter, courts generally agree they can hear claims over an agency’s failure to publish records in the Federal Register or post them in an online “reading room.” But at least one court of appeals has determined it cannot provide common-sense relief after concluding an agency has neglected its affirmative-disclosure obligations. That is, instead of entering an injunctive order that requires an agency to comply with its reading-room duties, the U.S. Court of Appeals for the District of Columbia Circuit has stated the available relief is limited to enjoining the agency, even prospectively, to producing records merely to the prevailing plaintiff. The argument for limiting the remedy in this way is that a more general publication order would go beyond curing an individual plaintiff’s informational injury. Two other circuits have found that position unpersuasive. 

At least one circuit judge, in concurrence, has cast doubt on the validity of policy-or-practice claims altogether, thereby opening the door to further erosion of courts’ remedial authority and the diversity of claims the statute envisions. Admittedly, the policy-or-practice theory is an odd outgrowth of a jurisprudential mootness exception that finds little purchase in the text of FOIA itself. But the theory is well-established and, more than anything, this shortcoming counsels in favor of Congress codifying such a claim to prevent its outright rejection by the courts. 

The government has also sought to leverage remedial authority against requesters, arguing that FOIA supplements a court’s inherent equitable power to enter “clawback” orders forcing requesters to return or destroy public records that an agency maintains were inadvertently disclosed. The D.C. Circuit has largely rejected the availability of “clawback”—albeit without addressing lurking First Amendment objections—yet the government is attempting to force the issue again.

Ultimately, Congress must carefully spell out a court’s remedial authority and the available causes of action a requester may bring under FOIA. By providing this textual clarity, Congress would reduce collateral disputes while reemphasizing that courts are meant to function as the “enforcement arm of the statute,” fashioning such relief as needed to fulfill the law’s promise of openness and transparency.

Restoring Meaningful De Novo Review

FOIA’s de novo standard is one of its most important features. Congress understood that agencies have institutional incentives to withhold embarrassing, inconvenient, or sensitive information. Hence the need for independent review. Unfortunately, many courts are not holding agencies to account. They have accepted agency efforts to “circumvent[]” de novo review by engaging in various types of “deference conservation,” especially in cases implicating national security or law enforcement matters.

This “deference” extends beyond legal questions—or the meaning of statutory terms, including the scope of FOIA’s exemptions—to factual matters or the application of law to facts. Yet these are the situations in which a fresh look by a judge would be most appropriate under FOIA. In one older district court case, Chemical Manufacturers Ass’n v. CPSC—which the government continues to cite in contemporary FOIA litigation—a judge decided he owed “considerable deference to the [agency’s] judgment as to what constitutes . . . ‘part of the agency give-and-take’—of the deliberative process,” at least so long as decision-making was still ongoing. That kind of solicitude for an agency’s evaluation of “what confidentiality is needed”—a kind of epistemic deference that relies on an agency’s supposed expertise—seemingly veers quite far from de novo review.

As others have explained, deviation from the de novo standard is also found in a court’s uncritical acceptance of declarations and Vaughn indices, as well as its reliance on substantive or procedural interpretations of FOIA that are structurally oriented toward affirming the government’s position. This problem of procedural deference extends to judicial doctrines like the presumptions of “good faith” and “regularity,” and the D.C. Circuit’s more controversial presumption of compliance with FOIA’s duty to segregate non-exempt material from otherwise exempt records.

Although informal deference—including the practice of allowing agencies repeated opportunities to supplement the evidentiary record with multiple rounds of summary judgment—may seem reasonable to some, disregard for the standard set by Congress has transformed much of FOIA litigation into a sort of box-checking routine in which attorneys and judges “focus[] on a dispute about the adequacy of the [agency’s] Vaughn index, rather than a dispute about the merits of the exemption claims themselves.” It is also unclear whether the practical erosion of de novo review can persist within the new “Loper Bright interpretive paradigm” of administrative law, with its emphasis on the role of independent judicial judgment. In any event, Congress needs to reinforce and empower judges to independently verify an agency’s factual claims whenever a requester raises genuine questions about search adequacy, exemption use, segregability, delay, or systemic noncompliance with the statute.

Accounting for Information Asymmetry

As the D.C. Circuit has explained, the “unique evidentiary configuration” of “the typical FOIA case ‘distorts the traditional adversarial nature of our legal system’s form of dispute resolution.’” Requesters face an “asymmetrical distribution of knowledge” in challenging agency action because “the agency alone possesses, reviews, discloses, and withholds the subject matter of the request.” Yet that “nearly impregnable defensive position” depends on the agency’s ability to meet its burden “to sustain its action.” But a nominal burden on the agency is not enough. Basic principles of fairness require agency arguments to be subject to genuine adversarial testing. The unavailability of discovery in most cases, total absence of an administrative record, and selective application of the federal rules of procedure and evidence frustrate an equitable playing field.

Congress should adopt procedures that reduce this imbalance. It could require agencies to proffer more detailed declarations, strengthen Vaughn index requirements, authorize discovery when there is a material factual dispute, encourage the use of FOIA trials, reinforce in camera review, and make clear that agencies bear a meaningful burden to justify the reasonableness of their searches, withholdings, foreseeable-harm determinations, and segregability efforts.

Conclusion

FOIA is often considered an administrative problem. Agencies lack resources, backlogs continue to grow, and requesters face interminable delays. Those concerns are real, but they do not fully account for the statute’s declining effectiveness. At its core, FOIA enshrines a judicially enforceable right of access. Congress vested federal courts with responsibility for independently evaluating agency action under FOIA. Decades of litigation and institutional cultural shifts have eroded that design. The result is a system in which courts too often validate agency secrecy rather than engaging in true independent judgment with an eye toward openness. Congress should act to restore the judiciary to the role it envisioned.

The amicus brief argues that the press has, for over a century, fulfilled a crucial role by reporting on newsworthy events from within the White House on behalf of the American people, and that the president's decision to ban news organizations from the White House based on the viewpoints expressed in their reporting violates the First Amendment.

On September 24, 2026, the district court granted the plaintiffs' motion for a temporary restraining order, requiring the government to restore the plaintiffs' access to the White House.

Status: Hearing on the plaintiffs' motion for a preliminary injunction scheduled for October 8, 2026 at 11 a.m in Courtroom 11. Members of the public may attend in person or by telephone. Toll free number: 833-990-9400. Meeting ID: 433360750.

Case Information: Cable News Network, Inc. v. Trump, No. 1:26-cv-03287 (D.D.C.).

But it is not just requesters who register high levels of frustration. As the number of requests increases almost exponentially year over year and the volume of records created and maintained by agencies balloons, the resources agencies are given—both human and technological—perennially fall far short of what is needed. On one side, FOIA offices struggle to communicate with requesters effectively even when communication would be mutually beneficial. And on the other side, those same offices face barriers in securing the cooperation of agency program officials in other units to search for and produce responsive records. Agency officials cite the near impossibility of the task at hand, and having been set up to fail, FOIA compliance contributes to a public perception of government incompetence.

In an attempt to support agencies in their FOIA processing, tracking, referring, communicating, and more, private companies have long offered various purported solutions. The leading product, FOIAXpress, has been used by many agencies for more than two decades, but competitors have cropped up like NextRequest and Armedia. Without doubt, these private companies are innovating in the FOIA processing space, including by trying to integrate across non-FOIA staff and by harnessing the power of artificial intelligence. Still, though, private products are adopted piecemeal, agency-by-agency, and do not integrate across the government. Moreover, the nature of private products has exposed government to potential security breaches. And finally, these products are expensive and offer many add-ons and customization options, making even the very same products function differently across agencies. For requesters, this means that each agency still has a bespoke requesting process.

At times, both Congress and the administration have recognized the need for greater ownership and consistency of technology in a centralized location in government. In the FOIA Improvement Act of 2016, Congress required the Office of Management and Budget to “ensure the operation of a consolidated online request portal that allows a member of the public to submit a request for records… to any agency from a single website.”

While this had the promise of taking a step in the direction of improving FOIA infrastructure, the implementation has proven wholly inadequate. FOIA.gov, the government’s attempt at moving toward a FOIA portal, acts mainly as a redirect site; that is, it provides a centralized area to find a link to each agency’s request submissions process. It does not, itself, allow a single submission site.

This is not the first attempt that has gone awry. The EPA spearheaded an earlier effort, FOIAonline.gov, a portal system that allowed other agencies to opt in to using it for request submissions on a single site and was used by twenty-two agencies at its height. But after the EPA experienced high costs in maintaining the system, it was decommissioned and shut down entirely in 2023.

Even at their zenith, none of these efforts have been adequately ambitious to produce the kinds of gains that real technology infrastructure might offer. Other countries have vastly outpaced us in this regard, and their models are instructive in the possible efficiency gains of technology platforms that serve the administrative needs of FOI professionals and the public alike.

During the height of Mexico’s transparency regime, Mexico’s then-extant Information Commission created a National Transparency Platform, an integrated portal system that out-imagines anything we have tried. As a basic matter, it allowed requesters to make a request to any governmental body at any level of government from a single portal, or even to make a request to multiple agencies at once. It also allowed the requester to log in and see all past requests made, with deadlines for responses clearly indicated, as well as all subsequent communication and any eventually released records. The portal served as a repository for all released records which (absent individual privacy concerns) were also posted publicly for anyone to search and review. All records required to be made affirmatively available were likewise housed on the portal.

But where it really shone was on the backend. The portal served as the internal processing management system for government entities. Each agency unit that processed requests (the equivalent of a FOIA office) could see all the pending requests and deadlines, could refer the request internally to the program or component of the agency that was likely to have responsive records—each of which had a designated contact for handling FOI requests—and could set an internal deadline for that program to respond to the FOIA office. The contact point in the program office, likewise, had their own login that provided a list of all referred requests pending and allowed them to build the record on their end.

Finally, the National Transparency Portal also facilitated communication between the agencies and the information commission; once the agency response was registered, it allowed requesters, with the click of a button and a sentence-or-two explanation, to administratively appeal the response to the commission. The commission then had access to the full administrative record of what happened prior to the appeal and could receive additional documentation through the portal, as well as request additional information from the agencies. At base, therefore, the portal looked different to each actor who had a touchpoint on a request by virtue of their login. It clearly managed that individuals’ workflow and facilitated retention of a full record of what happened.

Mexico is far from the only country to have developed such a deceptively simple sounding system, but one that radically transforms the FOI space. Brazil, for example, has a very similar one-stop-shopping portal system for their FOI space, as do Paraguay and Norway.

The U.S. needs a true FOIA portal. A real portal would create the infrastructure that would facilitate the possibility of something akin to full compliance with FOIA obligations. It would supplant a piecemeal system with a system designed for this purpose, built in house, maintained by government computer engineers, and consistent across agencies. It would streamline the workflow inside agencies, make clear who has the responsibility for the request at each stage, and create priorities for work based on deadlines. It would provide the public with one-stop-shopping for government information and lower the barriers to use by ordinary citizens. And it could create efficiencies for agencies, which would no longer need to invest FOIA budgets in procurement of third-party products for FOIA solutions.

The most exciting part of a true portal, however, is the nearly endless list of possibilities beyond improving the current request-and-response compliance rate. For example, a portal could facilitate release-to-one-release-to-all, a widely celebrated concept that once a record is released under FOIA it should be available publicly for anyone to access. The DOJ has supported the idea in pilot version previously, but it never became a reality. Indeed, implementing a release-to-one-release-to-all system would obviate the need for FOIA’s current requirement that agencies release publicly any record requested three or more times, a requirement agencies have found extremely difficult to implement. In fact, much more broadly, a portal creates an opportunity for meaningful affirmative disclosure regimes in which records are searchable government-wide and agencies have checklists of disclosure obligations with which they must comply.

Exemption application could be improved through the use of a portal as well. For example, a portal could collect the body of past agency decisions—both at the initial and the administrative appeal level—as well as adjudicated case law in one place. It also would contain all previously released records—whether at the administrative level or in response to litigation—thereby making plain when information is already public on a particular topic and facilitating a determination that further release would not harm a protected interest. Easy access to more definitive answers on the application of exemptions or other matters under the law might improve agency compliance in future cases.

A portal might also be used to facilitate data analysis, auditing, and other programmatic evaluation of FOIA offices. Patterns of violations—whether delay, overwithholding, or failure to engage in affirmative disclosure—could benefit from systematic solutions, and a portal could be used to facilitate identification of areas ripe for improvement.

Even more broadly, a portal could facilitate many of the other types of FOIA reforms that have been considered within the FOIA community. For example, it might allow tracking the personal responsibility attached to FOIA violations to particular employees in a way that could sanction flagrant noncompliance. It could facilitate improved judicial review by creating a full administrative record for the court and permitting easier in camera review of records. It could link to declassification efforts by the National Declassification Center or be integrated as a tool hosted by the Archives into the NDC’s work. It could be integrated into artificial intelligence solutions; imagine, for example, an AI tool that combed through agency records and identified records that are likely of interest to the public and can be affirmatively disclosed. And it may hold the potential for many more uses we could tailor the portal to as it developed over time. FOIA needs a portal now.

“Journalists covering the Pentagon play a crucial role in keeping Americans informed about the actions that the U.S. military and its leaders take in our name,” said Jake Karr, staff attorney at the Knight Institute. “These new restrictions represent a transparent attempt to cow independent, fact-based reporting and convert the Pentagon press corps into something more like a propaganda machine.”

Today’s brief argues that the government’s unprecedented restrictions violate the First Amendment’s public forum doctrine because they are impermissibly motivated by a desire to suppress critical journalism and because they undermine the very purpose of allowing journalists to access the Pentagon in the first place—to facilitate accurate and timely reporting on matters of war and peace that informs and serves the public. It urges the court to roll back these restrictions and restore the Pentagon press corps’ ability to gather and report the news without fear or favor.

“The administration is trying to prevent journalists from publishing reports about the military that it simply doesn’t like,” said Katie Fallow, deputy litigation director at the Knight Institute. “Especially in times of war, the public needs accurate, timely, and independently sourced information that only a free press can provide.”

The Department of Defense issued the interim policy that is the focus of this appeal following a ruling by the district court that the original policy restricting reporters’ access to the Pentagon was unconstitutional; the government has not challenged that ruling.

The Knight Institute filed a similar amicus brief earlier this month in a related case also challenging the interim policy.

Read today’s amicus brief here. 

Read more about the cases—New York Times v. Department of Defense (NYT I and II)—here. 

Lawyers on the case include Katie Fallow, Jake Karr, Jameel Jaffer, and Alex Abdo for the Knight First Amendment Institute.

Today, the Knight Institute also announced that it will support CNN, MS NOW, and Politico in their lawsuit against the Trump administration’s White House press ban. Read more here.

For more information, contact: Lorraine Kenny, lorraine.kenny@knightcolumbia.org. 

The following can be attributed to Jameel Jaffer, executive director, Knight First Amendment Institute at Columbia University

“We’re glad to see that these news organizations are suing. The president’s decision to bar them from the White House for the content of their reporting is a flagrant violation of the First Amendment. It’s an attack both on press freedom and the public’s right to independent information about the government. We will support these news organizations in court.”

The Knight Institute has supported the Associated Press in its challenge to its exclusion from the White House, filing three separate amicus briefs over a period of 18 months. The Institute has also supported The New York Times in its challenge to restrictions on access to the Pentagon. Earlier, the Institute successfully sued President Trump over his blocking of critics from his Twitter account, which the Institute argued was a “public forum” for First Amendment purposes. 

The following can be attributed to Jameel Jaffer, executive director, Knight First Amendment Institute at Columbia University

“The First Amendment prohibits the president from punishing journalists because he doesn’t like their coverage, just as it prohibits him from punishing universities because he doesn’t like the courses they offer, or from punishing law firms because he doesn’t like the clients they represent. With so many courts having ruled against him on exactly this point, you’d think President Trump would have learned this lesson by now. If President Trump means to expel these news organizations from the White House press pool, his action is doubly unconstitutional because the press pool is a ‘public forum’ under the First Amendment, which means the president can’t exclude journalists from it on the basis of their viewpoints.”

For more information, contact: Lorraine Kenny, lorrraine.kenny@knightcolumbia.org. 

These twin crises have plagued—and been exploited by—Democratic and Republican administrations alike, demonstrating that the challenge is not partisan, but structural.

It is past time for reform. Here are four ways to start.

1. Congress must stop deferring to the executive branch on classification

One of the classification system’s shortcomings is that it is governed almost exclusively by executive order. (Nuclear weapons information governed by the Atomic Energy Act of 1954 remains the most notable exception.) Decades of overclassification show that leaving secrecy rules solely to executive discretion leads to too many secrets being created, with too few opportunities for oversight by the judicial or legislative branches.

Nearly three decades ago, the bipartisan Moynihan Commission on Protecting and Reducing Government Secrecy studied this very problem. It found that the “culture of secrecy” was so deeply entrenched in the executive branch that Congress must enact statutory reform and pass legislation mandating that records only be classified if there was a demonstrable need, along with other improvements.

That recommendation was largely ignored, and the culture of secrecy remains deeply embedded within agencies. Congress must finally take up the Moynihan Commission’s recommendation to set classification rules.

2. Standardizing definitions and incentivizing individual accountability

Most classification decisions are subjective. It is not unusual for the State Department to approve a document for declassification, only to be overruled by the CIA, which insists the material must remain secret. This is caused by a lack of objective standards across (and within) agencies that leads to reflexive secrecy as a matter of routine.

Congress can fix this in two ways:

• Clearly define classification standards in statute: Legislation should establish an interagency task force, led by the Information Security Oversight Office (ISOO) at the National Archives and Records Administration (NARA), charged with strictly defining the vague phrase "damage to the national security." The Classification Reform for Transparency Act, a bipartisan bill introduced in 2024, introduced something similar, and its proposals should be built upon.
• Incentivize leadership and penalize abuse: Legislation should compel agency leadership to proactively fix classification. This could include imposing financial penalties to be incurred by senior leadership for agencies that habitually overclassify information, and could be coupled with a financial award program for agency personnel who successfully lodge good-faith classification challenges.

3. Quantify the size and cost of the problem, then impose a 10% "declassification tax"

The biggest practical hurdle in reining in overclassification is that there are no reliable metrics on the size and cost of the classification system.

The numbers problem is so bad that the last fiscal year for which we have data on the number of original and derivative classification decisions is 2017, when ISOO reported 58,501 original classification decisions and 49 million derivative classification decisions. Since then, ISOO has not reported government-wide classification decisions because agency data was notoriously poor and inconsistent, making the data effectively useless and leaving policymakers without a metric to evaluate secrecy reform ever since.

New legislation should institute:

• Mandatory audits and standardized metrics: Agencies that classify information should adopt standardized data tracking for original and derivative classification decisions, including accounting for the explosion of derivative decisions generated by new technologies on classified networks, to be reported annually in public-facing audits. This, coupled with a clearly defined statutory basis for classification, should improve classification decisions off the board.
• The 10% declassification tax: Part of this audit must address the ever-growing costs of classification. Reforms should require every agency to calculate its total annual classification costs—accounting for the costs of security clearances, SCIF maintenance, classified IT infrastructure, such as SIPRNet and JWICS, and other relevant metrics. Once tabulated, 10% of that publicly-reported total should be invested in agencies’ declassification programs. This idea has been broached before, and it should be put into law.

4. Empower NARA to declassify other agency records and double its budget

Legislation must also address the primary bottleneck in historical declassification: the endless interagency referral loop. Currently, these releases stall for decades in large part because NARA does not have the authority to declassify other agency records in its possession and must refer files back to the originating agency, which in turn may refer portions of those requests to numerous other agencies, which must all conduct their own, time-consuming reviews.

To fix this Congress must:

• Grant the Archivist of the United States and the National Declassification Center the authority to declassify records in NARA's custody that are 25 years or older. By stripping originating agencies of their power to keep historical files secret, the statute will eliminate the referral loophole that keeps history hidden.
• Double NARA's Budget. NARA is already woefully underfunded. For the agency and its components like ISOO and the NDC to be successful, it needs an influx of funds for staff and to acquire modern technology capable of working with all of the different kinds of classified records that intelligence and defense agencies possess and transfer to it. This will have the added benefit of boosting NARA’s other mandates, including ensuring that agencies are properly creating and preserving the records they are required to retain.

Critics of congressional involvement in classification will point to the disaster that was the Kyl-Lott Amendments of the late 1990s, which made declassification of historical nuclear-related documents even more cumbersome. While it's clear the executive branch needs legislative guidance to improve overclassification, any new legislation must learn from these amendments’ mistakes and prioritize clear, quantifiable standards over vague and retroactive mandates.

Finally, classification reform is not a stand-alone issue. It overlaps with records management and transparency requirements, and fixing one issue while ignoring the others will ultimately fail. For example, Congress must also account for senior officials who willfully destroy records to keep them from the public and consider punitive measures to deter this growing threat. To do so, it could make intentional records destruction (or overclassification) a disqualifying factor for running for public office or have such offenses result in the immediate revocation of security clearances.

Reducing overclassification and restoring meaningful oversight of classification decisions requires moving beyond executive orders. Codifying metrics, establishing a declassification tax, narrowing the definitions of damage to national security, and granting NARA authority to end forever secrets, will shrink the universe of government secrets and restore public trust.

“Today’s decision leaves Americans’ most sensitive information open to search at the border without any suspicion at all,” said Scott Wilkens, senior counsel at the Knight First Amendment Institute. “Our phones hold our private thoughts and associations, photographs of our family and friends, and a log of our nearly every movement. The First Amendment should require the government to get a warrant before searching them. We’re disappointed the court declined to recognize that.”

Today’s decision involves a criminal case, United States v. Alisigwe, in which the government relied on evidence obtained from two warrantless searches of the defendant’s cell phone at the border. In November 2023, the district court denied the defendant’s motion to suppress the evidence. The Knight Institute and RCFP’s amicus brief before the Second Circuit pointed to documents obtained by the Knight Institute through FOIA litigation in Knight First Amendment Institute v. Dep’t of Homeland Security and also discussed the burdens these searches place on journalists, whose electronic devices contain sensitive newsgathering information, including the names of confidential sources. The brief argued that the border-search exception to the Fourth Amendment’s warrant requirement does not apply to searches of electronic devices and urged the court to conclude that the First and Fourth Amendments require the government to obtain a warrant before searching a cellphone at the border. The Second Circuit rejected these arguments in today’s ruling.

In March 2025, the Knight Institute’s Wilkens argued before the Second Circuit. 

Read today’s decision here.

Read more about the lawsuit, United States v. Alisigwe, here.

Lawyers on the case include Scott Wilkens, Alex Abdo, and Jameel Jaffer of the Knight First Amendment Institute.

For more information, contact: Lorraine Kenny, lorraine.kenny@knightcolumbia.org.