When the Obama administration issued its 2009 Open Government Directive, it promised a new era of digital transparency: agency data portals, machine-readable records, and freedom of information requests processed at network speed. Fifteen years later, that vision has largely collapsed under its own contradictions. FOIA backlogs have reached record highs, classification decisions number in the tens of millions annually, and journalists routinely wait years for documents that agencies could technically deliver in seconds.

The paradox is worth sitting with. The infrastructure for radical transparency exists—cloud storage, full-text search, automated redaction tools—yet the practical experience of extracting information from government has become measurably worse. Something in the institutional response to digital abundance has produced defensive contraction rather than democratic opening.

For news organizations, this matters beyond the frustration of individual reporters. Access to government information constitutes the raw material of accountability journalism, and its progressive constriction reshapes what stories can be told, which powers can be checked, and how citizens understand the state acting in their name. The technology promised to democratize access; the political and bureaucratic response has, in important ways, done the opposite.

The Freedom of Information Act was drafted in 1966 for a filing-cabinet bureaucracy, and its statutory architecture still assumes that document retrieval requires physical labor. Agencies invoke this legacy framing to justify processing times that have grown dramatically over two decades, even as the underlying records have become vastly easier to search and produce.

The numbers tell a coherent story of institutional decline. Federal FOIA backlogs have roughly tripled since the mid-2000s, with some agencies reporting median processing times measured in years rather than the statutory twenty working days. Complex requests—precisely the ones journalists need for investigative work—routinely take five to seven years, effectively neutralizing their news value.

Resource starvation explains part of this dysfunction. FOIA offices are chronically underfunded relative to request volume, and no political constituency exists to advocate for larger transparency budgets. Agencies have little incentive to invest in a function that primarily generates institutional risk. The result is a slow-rolling capacity crisis that appears procedural but is fundamentally political.

Exemption expansion compounds the delay problem. Agencies increasingly invoke deliberative process, law enforcement, and personal privacy exemptions with maximalist interpretations that courts occasionally reject but rarely reverse at scale. Fee waivers for journalists, once routine, now require documentation battles that consume months. Each friction point functions as a filter, screening out all but the most persistent requesters.

The cumulative effect is a system that satisfies formal legal requirements while frustrating substantive access. Requests are logged, acknowledged, and eventually answered—often after their relevance has expired. Compliance becomes a performance rather than a practice, and the statutory right of access becomes something closer to a lottery ticket.

Takeaway

Legal rights without institutional capacity are ceremonial rather than substantive. A transparency regime that formally exists but functionally fails may be more corrosive to public trust than open opacity would be.

The volume of information classified by the U.S. government has grown at a rate that vastly exceeds any plausible growth in actual national security equities. The Information Security Oversight Office has repeatedly documented tens of millions of classification decisions per year, with derivative classification—stamping secrecy onto documents that reference already-classified material—accounting for the overwhelming share.

This expansion reflects an incentive structure rather than a threat environment. For classifying officials, the professional cost of over-classification is essentially zero, while the cost of under-classification can be career-ending. Rational actors within this asymmetry classify aggressively, and the resulting corpus of secret information grows regardless of whether the underlying material warrants protection.

The categories themselves have proven remarkably elastic. Information about drone operations, surveillance programs, diplomatic assessments, and even public statements by foreign officials have all been swept under classification regimes. The Reagan-era shift that permitted classification of information whose disclosure might cause undefined harm—rather than requiring specific identifiable damage—remains substantially in place across administrations.

For journalism, this creates a peculiar epistemological problem. Reporters covering national security often know things they cannot report because official confirmation would require declassification, while their sources cannot speak on the record without risking prosecution under aggressively interpreted statutes. The public conversation about foreign policy, intelligence, and military affairs proceeds atop a substrate of enforced silence.

Reform efforts have consistently stalled because no political coalition benefits from narrowing classification authority. Presidents inherit the apparatus and quickly discover its usefulness. Congressional oversight committees are read into secrets they cannot discuss publicly. The classification system persists not because anyone defends its current scope but because no one has the political capital to shrink it.

Takeaway

When the personal cost of caution is zero and the cost of disclosure is high, institutions will systematically over-restrict information regardless of stated policy. Incentive structures shape outcomes more reliably than principles.

The theory behind digital transparency was elegantly simple: if records exist in searchable databases, publication becomes a matter of policy rather than logistics. Data.gov, agency dashboards, and open-format mandates would route around the gatekeeping problem by making disclosure the default state. Two decades of implementation have shown why this theory misunderstood the actual bottleneck.

The constraint on transparency was never primarily technical. Agencies that resist disclosure through FOIA have proven equally capable of resisting disclosure through digital channels, simply by controlling what data gets published, at what granularity, with what caveats. Proactive disclosure programs have often become vehicles for releasing information agencies want public while providing cover for the continued withholding of information they do not.

Platform intermediation has introduced new complications. Government records increasingly live in commercial cloud environments, communication happens across encrypted messaging apps that officials use to route around record-keeping requirements, and administrative decisions get made in Slack channels and text threads that were never designed for archival preservation. The digital transition has, in important ways, made government activity less legible rather than more.

Automation has also cut against transparency in unexpected ways. Machine-assisted redaction tools allow agencies to process large document sets while applying exemptions more aggressively and consistently than human reviewers could. Algorithmic decision-making by government agencies introduces new categories of consequential activity that exist below the threshold of traditional record-keeping. What gets automated often becomes what gets hidden.

The lesson is not that technology cannot serve transparency, but that tools inherit the politics of their deployment. Where institutional will exists, digital infrastructure enables remarkable openness. Where it does not, the same infrastructure enables sophisticated new forms of opacity. Technology amplifies existing dispositions rather than replacing them.

Takeaway

Tools do not have politics of their own; they carry the politics of the institutions that deploy them. Building better transparency infrastructure without addressing institutional incentives produces better-organized secrecy.

The expansion of government secrecy amid digital abundance is not a technological failure but an institutional one. The infrastructure works; the incentives point the wrong way. Understanding this distinction matters because it changes what reform requires.

For journalism, the implications are strategic. Investment in litigation capacity, source protection, and collaborative investigative networks may yield more accountability leverage than continued reliance on formal access mechanisms. The organizations extracting real information from government today have largely built their own workarounds to a system that no longer functions as designed.

The deeper question for democratic information is whether societies can sustain accountability journalism when the formal architecture of access has decayed while the informal architecture remains contested. That answer will be shaped less by transparency law than by the political will to fund, defend, and enforce it.