Federal and state governments have leaned on technology companies to provide an ever-increasing range of core government activities—from the use of algorithmic decision-making in the criminal justice system to powerful surveillance technologies to providing the government’s core information infrastructures. As privately owned companies continue to become intricately involved in many areas of governance, courts should recognize a First Amendment right of access to certain information held by these companies in order to ensure that citizens’ long-enjoyed right to obtain information about their government and to hold power accountable remains viable in an era in which private actors exert significant control over our civil liberties. Expanding the First Amendment’s right of access would be a radical and robust measure and one that would directly disrupt the disturbingly asymmetrical power dynamics exerted by private entities. By ensuring citizens’ right to knowledge and self-government in the digital age this measure would ensure that the government could not obfuscate civil rights and civil liberties violations or evade general accountability by operating through privately owned proxies.

The right of access as a means of ensuring government transparency has strong roots in the American legal tradition. In 1961, legal scholar Alexander Meiklejohn argued that democracy depends on a citizen’s ability to obtain information necessary for self-governance and that the First Amendment is intended to ensure this structural protection for democracy. Noted scholars like Thomas Emerson have found support for this conception in the writings of James Madison, author of the First Amendment. Madison famously asserted that “[a] popular government, without popular information or the means of acquiring it, is but a prologue to a farce or a tragedy; or perhaps both.” Supreme Court Justice William Brennan also referred to this Madisonian principle in a famous right of access opinion that highlighted the right’s “structural role . . . in securing and fostering our republican system of self-government.” Ultimately, this American right is enshrined in the common law, the Constitution, and various laws including the Freedom of Information Act (FOIA) and its state-level equivalents.

As our democratic institutions have become more complex, these rights have been expanded over time to meet the challenges of increasing opacity. In the 1950s, as the number of executive agencies grew, amassing unprecedented power with little direct accountability to the public, Congress passed the FOIA over a presidential veto in order to ensure freedom of information. As recently as 2016, Congress echoed the importance of this view by passing an amendment to the FOIA that “emphasizes in no uncertain terms . . . that the balance between the American people’s right to know about their information and the government’s right to keep a secret shall always be balanced in favor and presumed to be the American people’s right.” In championing this bill, Jason Chaffetz, former chairman of the House Committee on Oversight and Government Reform, specifically highlighted the overuse and misapplication of FOIA Exemption 4, which permits the government to withhold confidential business information.

Today, we are again at a moment when government has become even more complicated and harder to hold accountable, as privately owned technology companies become inextricably intertwined with federal, state, and local government actions. This corporate dominance is born out of an ongoing trend of private intermingling with the administrative state. In 2002, Martha Minow, the former dean of Harvard Law School, wrote that “schools, prisons, welfare agencies, and social service programs” are increasingly undergoing a process of privatization. But what was once a nascent trend has now grown into the status quo. As of 2015, over 40 percent of the federal workforce was made up of contractors, many of which are primarily technology companies. And the services that these technology companies are providing are ever more core to our democratic order and are thus intricately woven into the most intimate parts of our lives. Technology companies’ algorithms determine “whether you get a job interview, how much credit you access, and what news you see” as well as the level of your health care benefits or whether you will be charged for a given crime.

As their structural power grows, these entities evade public scrutiny in how they carry out government functions, a situation that carries a potential for enormous negative consequences. Little is known about how these companies collaborate with and advise the government in part as a result of their fervent insistence that their actions remain in the private realm. Companies that contract with the government are afforded significant privacy in carrying out government tasks through various trade secret and confidential business information exemptions to public records laws, preventing public accountability over government processes and power. In some cases, the companies themselves are left in the dark about how their technology is employed by the government. In response, this article suggests that the right of access should reach beyond algorithms to all areas of corporate governance through which government functions are influenced, controlled, or dominated by private technology companies.

This proposal echoes a solution to a Gilded Age debate over whether the principles of democracy demand that transparency rights be extended to private industry. Over a century ago, philosopher John Dewey first argued that as the Industrial Age introduced new, complex, and hidden technologies into society, better education about these systems was necessary to the functioning of democracy. Dewey initially proposed this as a pedagogical theory in his work School and Society, and in subsequent years, a small group of scholars revived it and named it “technological transparency.”

Borrowing from Dewey, this article argues that similar technological transparency is necessary in what some have called our new Gilded Age, as technology companies have grown to occupy a quasi-governmental role in society, directly impacting individuals’ civil rights and civil liberties and controlling other key government functions. As we shall see, these companies have become not only the primary influencers over online speech but also integral agents in the functioning of the criminal justice system, the military, and even international diplomacy. A right of access is necessary to balance these structural changes in institutional power. And while other proposals to increase tech companies’ accountability through taxes, regulations, antitrust enforcement, or fiduciary obligations are helpful, an expanded right of access remains the most basic and most necessary fix that could assist all other reforms.

To make the case, this paper will first describe how technology companies are increasingly acting as quasi-government institutions. Next, it will define the roots and parameters of the right of access as well as a short explanation of Dewey’s original notion of technological transparency. The final section will take up the case that particular democratic principles especially require considering expanding the right of access to technology companies.

I. Silicon Valley Companies: The Call for Transparency in the New Towns

Nobody seriously doubts that the titans of “Big Tech” have begun to dominate our lives. Over the past two decades, several Silicon Valley companies have become extremely powerful in controlling speech. Reflecting that reality, Facebook decided this past year to create a new “Supreme Court” of speech, a body of forty salaried judges who will work part time for three years and who will have the ability to rule on Facebook’s hard cases to determine whether or not to delete certain information. Academics like Kate Klonick and Jack Balkin have written about the growing power these media giants wield over speech and have suggested various accountability measures such as treating these new governors as information fiduciaries with certain obligations. However, the power these companies exert extends well beyond the sphere of speech. Amazon, which surpassed a trillion dollars in market value (second only to Apple) in 2018, currently offers a variety of tools to local law enforcement agencies around the country and has recently captured 46 percent of online shopping in addition to delivering packages and acting as a credit lender, a producer of content, and a leading provider of cloud server space. Facebook has established a market value of more than $600 billion and has faced scrutiny for its improper influence in the 2016 election and its more recent decision to create its own currency. Google, one of the world’s greatest aggregators of personal data, has become heavily involved in providing tools for the Department of Defense and has access to highly granular information about our daily lives.

To a large extent, these companies are acting as quasi-sovereigns—through their sheer scale and breadth—wielding power over multiple important aspects of our lives; and just likeany other government entity, they should be held accountable by making their information public. For instance, it is well acknowledged among scholars, lawyers, and journalists that through algorithmic choices, technology companies are transforming public life in completely novel ways compared to other industries. Some thinkers predicted the emergence of this new form of power in the internet era; for instance, John Perry Barlow, the self-dubbed internet pioneer, clearly presaged that a new “sovereignty” would arise on the internet. And while academics like Jeffrey Sachs have long warned the public about the rise of “corporatocracy,” Silicon Valley companies have several unique properties that demand a need for transparency rights to apply to them.

A. Technology Companies That Exercise Government-Like Control and Functions Should Be Subject to the Same Accountability Measures

Silicon Valley companies wield power over the most intimate parts of our lives, including having the ability to influence our identities and customs, a notable capacity at a moment when our allegiances to cultural traditions, religion, and unified visions of a nation state are waning. In some cases, the government is relying on external companies to also control sensitive identifying information like Social Security numbers and medical records, and the government has even relied on private vendors to issue government-backed credentials. In his book The Black Box Society, Frank Pasquale has written aptly about the control Big Tech’s algorithms wield over our lives, describing how “credit raters, search engines, major banks, and the TSA take in data about us and convert it into scores, rankings, risk calculations and watch lists” that further categorize us in ways that shapes our identity. Recent news articles have also observed secret Facebook groups that have provided forums for police officers and Customs and Border Protection agents around the country to cultivate extremist mindsets showing the power of these tools to – if not intentionally then passively – incubate certain persuasions and group thinking in the ways that quasi-government actors often do.

Second, these companies wield power over the most basic and traditional functions of government. These range from influencing the development of transportation systems to restructuring banking functions (through apps like Venmo) to hosting public information and serving ads that influence elections to providing for the nation’s physical safety. As an example, several Silicon Valley companies are working closely with Department of Defense and local law enforcement agencies to provide surveillance and technological tools like facial recognition systems, license plate readers, and police body cameras. Amazon Rekognition, to name one such tool, is used by the CIA as well as state governments to detect objects, scenes, and faces, even though the tool has been found to employ inappropriate gender and ethnic biases. Palantir’s Gotham service, to name another, is used by three hundred California cities (collectively home to about 7.9 million people) through the Department of Homeland Security to assist local law enforcement via the use of surveillance tools that supposedly attempt to predict crimes. These tools have in some ways supplanted the most essential government function of police power - the capacity of the state to enforce laws and regulate behavior.

Third, unlike other private entities, these companies are unique in that they provide multiple government functions at once, making transparency simultaneously more important and more difficult. Tech companies can build tools and programs for schools, banks, aeronautical companies, and the police all at the same time. Amazon, for instance, provides storage for many government records in “the cloud,” offers law enforcement tools, and provides credit. Facebook has experimented with building planes and drones, exerts powerful force over our elections, and is exploring the creation of its own cryptocurrency. As New York Times technology reporter John Roose has observed, the companies’ operations are “sort of fractal . . . there are ten different parts of [the company] that you don’t understand at all. . . . And so that makes it very hard to create a unified theory of [these companies] and how [they] work.” The need for accountability increases in cases like these where institutions carry out multiple government functions at once.

It is also clear that efforts to ensure accountability and transparency from these companies have not kept pace with their expansion into areas traditionally controlled by, or at least regulated by, government. Several companies offer services allowing users to pay each other online, Amazon is a credit lender, and Facebook plans to create a currency, but none of these companies are regulated in the way that financial service companies are. They build tools to assist in posting job advertisements, scouting talent, and filling vacant positions, but the algorithms they create are not subject to employment laws. They assist law enforcement with surveillance and incarceration but are not treated like police.

Last and perhaps more symbolically, tech giants have also taken on the emblematic role of quasi-state actors. While presidents have often conferred with powerful corporate actors over particular subjects, it has seldom been the case that a single industry has held sway over so many different facets of politics from trade to defense and even, some say, exerting ideological sway over the decisions of a sitting president. Beyond the national arena, Silicon Valley companies have taken a seat at the international stage. In December 2015, the EU launched a multi-stakeholder Internet Forum where representatives from Facebook, Microsoft, Twitter, and YouTube sat alongside ministers from EU states to begin negotiating a voluntary code of conduct on terrorist content and hate speech online. Representatives of several foreign powers have made the journey to Silicon Valley to speak with the modern sovereigns; an ambassador from Denmark arrived in Silicon Valley with a letter from the Danish queen ; and Germany, France, and Slovakia have all recently appointed “digital ambassadors.”  It is safe to assert that some important international bodies now regard Big Tech concerns as quasi-sovereign entities.

B. Why Technological Transparency Is the Solution

Given Silicon Valley’s ever-growing influence over government, various scholars have discussed how to neutralize this asymmetrical power structure by arguing for various accountability measures. Much of this conversation is concerned with regulating Silicon Valley’s power over speech. For instance, Professor Edward Lee has suggested a “hybrid agency” that would oversee various takedown requests. Professor Jody Freeman has argued—in an argument that parallels one I have made —that the companies’ roles as information agencies should trigger certain legislative restrictions. Perhaps the most alluring model is that of Jonathan Zittrain and Jack Balkin, who assert that social media companies are information fiduciaries, like lawyers and doctors, and therefore should be held to higher standards of loyalty and care around speech as well as privacy. Some have persuasively argued for the return of a strong antitrust regime.

While many of these models are convincing, applying any of them alone without a foundational right of access would be incomplete. Many of these models even admit that more transparency is necessary for their framework to work. Unlike other models, a transparency framework would hold Silicon Valley companies accountable for their control of information that extends far beyond the realm of speech forums. The extension of the First Amendment right of access to certain corporate proceedings and records of technology companies would uniquely account for these companies’ increasingly unchecked influence over various realms of government in a way that the aforementioned proposals do not. For instance, while the information fiduciary model is limited to the “special relationships between companies and the people,” the right of access is made available to all persons, not only those who have opted to agree to a site’s terms of service, in the same way that the FOIA is available to all citizens. Moreover, demanding a more robust transparency framework speaks to the public distrust in many of the tech giants, particularly after repeated revelations of corporate misconduct. Where the fiduciary model is based on “trust,” the right of access model appropriately matches the skepticism that tech giants deserve to be confronted with as they wield their (ever-increasing) asymmetrical power in complex and secret operations and have been shown to abuse the public trust. If it was ever appropriate to accord companies like Facebook and YouTube the benefit of the doubt in such weighty affairs, those days are surely in the past. Unlike the information-fiduciary model that could be used to shelter abusive and manipulative corporate behavior, a robust transparency would not have the same limitation. As Justice Brandeis famously said, “sunlight is the best of disinfectants.”

II. Roots of the Right of Access

The right of access is rooted in the philosophy that because citizens are the ultimate sovereigns, they must be able to access information necessary to hold government accountable. As David Arcadia notes, this principle “underlies the First Amendment’s structural role as a facilitator of democratic control.” While the First Amendment is thought to primarily protect speech, scholars have often considered the right to free speech as a complementary component to the Amendment’s “larger commitment to transparency.” As First Amendment scholar Robert Post has stated, democratic legitimization “requires that government action be tethered to public opinion. . . . It is for this reason that First Amendment coverage presumptively extends to all communications that form public opinion” and by contrast “[a] state that controls our knowledge controls our minds.” Or, as free speech scholar Vincent Blasi has argued, the general populace must first be privy to the behavior of its rulers in order to judge them. In essence, we need access to information that impacts our lives in order to make meaningful decisions about democracy.

The tenets underlying the right of access are rooted as far back as the seventeenth century in the English judicial system, predating the formation of the United States. As the Supreme Court has noted, under English common law, public access to court proceedings was “the rule in England from time immemorial.” English courts consistently held that one could access government records where such access would benefit the public. Under that reasoning, a member of the public was permitted to inspect government documents and claim that right in situations where records would benefit litigation. Eventually, British courts would construe access to information even more broadly. For instance, some British courts found that, once a document was entered into evidence before a judge, individuals were granted a presumption of access to it.

Early American courts quickly adopted and expanded this practice. In the early United States, courts held that the attorney general could bring a suit to access non-judicial records on behalf of the public; some jurisdictions within the United States recognized an even broader right of access by allowing the citizen to bring the case in his or her own name. Eventually suits became allowed in most circumstances where litigation might expose a public wrongdoing or help monitor government functions. In 1978, in Nixon v. Warner Communications, Inc., the Supreme Court firmly held that the right of access applies to both public records and court records.

Today, the First Amendment’s broad and presumptive right of access is hard to overcome even where privacy and proprietary interests are asserted. Parties advocating for withholding must show specific negative effects—in particular, a privacy harm to an individual or the disclosure of confidential business information. For instance, access to records can be restricted in cases involving a rape, family dispute, or sexual crime in order to protect the privacy of the victims and especially the minors involved. But even in those highly invasive cases, courts have found that redactions or other remedies may be sufficient.

Similarly, courts have routinely found that corporate parties asserting “commercial interests” cannot overcome the presumptive right of access. In a 2005 federal court case, the defendant, a chain of grocery stores argued that sealing was necessary because disclosure of business records would put the company at a “competitive disadvantage” and harm their future negotiating position with labor unions. The court rejected these arguments, explaining that it would not “speculate” about how the company might be affected “at some point in the future in an unidentified labor dispute.” Similarly, in a case involving Apple, another federal court struck down assertions that records must remain under seal where they “reflect internal Apple processes and deliberations that Apple regards as highly confidential.” The court wrote, “the mere fact that the publication of records may lead to a litigant’s embarrassment or exposure to further litigation is not sufficient to meet the ‘compelling reasons’ standard.”

While courts will sometimes concede to trade secret justifications for withholding documents, even those concerns may be outweighed where the public interest mitigates in favor of disclosure. In a recent case I litigated to access court records involving Facebook’s “friendly fraud” program (aimed at taking advantage of children who were unaware of the charges imposed by platform); the court ordered disclosure of the records, overruling Facebook’s claims that disclosure “would put Facebook at an unfair competitive disadvantage in dealing with its partners and competitors.” The court found that “Facebook provided no specific support for the argument that revenue figures from nearly five years ago would impact current partnerships or provide undue advantage to its competitors” and “[b]y contrast, this information would be of great public interest.”

Relatedly, the Supreme Court has suggested that the right of access may also extend to other nontraditional “institutions,” if the effect of the extension would encourage trust of government among citizens. In Richmond Newspapers, Inc. v. Virginia, the seminal case on access to court records, Chief Justice Warren Burger, writing for the majority, asserted, “People in an open society do not demand infallibility from their institutions, but it is difficult for them to accept what they are prohibited from observing.” This key statement appears to imply that the right of access, which has been continually expanding over time, could apply to other, nongovernmental institutions, such as technology companies, particularly where those institutions are closely linked to providing government functions.

In determining whether to expand the First Amendment right of access to an untested area, courts today apply the well-known “experience and logic” test. The two-part test asks (1) “whether the place and process have historically been open to the press and general public” and (2) “whether public access plays a significant positive role in the functioning of the particular process in question.” If both of these questions are answered in the affirmative, the constitutional right of access applies. The right may be overcome only if denying access is necessary to serve a “compelling government interest” and if the limitation of the public’s right of access is “narrowly tailored to serve” that interest.  

Applying this test to technology companies would be admittedly difficult. Most obviously, there is no historical precedent for directly expanding the right of access to institutions outside government. However, several factors lean in favor of such expansion. First, and most important, courts have long underscored that “[h]istory has taught us that secrecy and lack of access to information allows for abuses of power” and that the right of access aptly applies in circumstances of consolidated power. Additionally, the right of access is commonly applied to circumstances where other avenues for accountability are lacking. Therefore the fact that technology companies’ obfuscation around data collection cannot be improved by other regimes like notice-and-choice or even more robust comment procedures leans in favor of expansion. Last, courts have also expanded the right of access to institutions outside the judicial system, such as administrative bodies, particularly those that have a large impact on the public’s rights, similar to technology companies. In considering whether to apply the right to these new sovereigns, we may consider the extent to which these companies exercise unusual influence over the citizenry, have a pronounced history of secrecy, and lack public accountability.

In arguing for this expansion, we may also return to the writing of John Dewey, who, like other American progressives at the turn of the twentieth century, argued that technological transparency was necessary to a healthy democracy. Reflecting on the rapid industrialization taking place in America Dewey argued for more transparency in order to expose how emerging technologies controlled political decisions occurring every day. In his 1899 book School and Society, Dewey wrote how the drastic change from rural to industrial life had obscured many quotidian but elemental processes in everyday life:

Instead of pressing a button and flooding the house with electric light, the whole process of getting illumination was followed in its toilsome length from the killing of the animal and the trying of fat to the making of wicks and dipping of candles. The supply of flour, of lumber, of foods, of building materials, of household furniture, even of metal ware, of nails, hinges, hammers, etc. was produced in the immediate neighborhood, in shops which were constantly open to inspection and often centers of neighborhood congregation. The entire industrial process stood revealed.

While citizens of rural America could easily understand the technologies that underpinned their daily lives, during the Industrial Revolution that knowledge quickly disappeared. Dewey exclaimed, “How many of the employed are today mere appendages to the machines which they operate!” It seems that similar exclamations about the “smart” technology used by government in ways that directly impact our lives and safety could be made in the present day as well.

Though it may be hard to envision courts expanding the right of access to private actors, Dewey foresaw the potential for this a century ago when he argued for holding public and private actors accountable. In Dewey’s later work The Public and Its Problems, he theorized that certain powerful institutions benefited from obfuscation and that, due to the key roles they play in governance, these private elite actors must be checked by some transparency. “Representative government must at least seem to be founded on public interests as they are revealed to public belief,” wrote Dewey. Worried about concerns over propaganda, he continued, “[t]he smoothest road to control of political conduct is by control of opinion.”

As long as interests of pecuniary profit are powerful, and a public has not located and identified itself, those who have this interest will have an unresisted motive for tampering with the springs of political action in all that affects them. Just as in the conduct of industry and exchange generally the technological factor is obscured, deflected and defeated by “business,” so specifically in the management of publicity.

It is not enough, Dewey would seem to argue, to limit the public’s right to know to strictly traditional governmental actors, since government is not the only sector of civil society that can be held responsible for the nation’s actions. For a democracy to stay healthy, powerful private actors—especially corporations—must operate with a degree of transparency.

III. Expanding the Right of Access

Considering the quasi-governmental role of Silicon Valley companies, various state action theories may help extend the right of access to these entities. While select courts have concluded that private companies like Facebook do not qualify as state actors and are not bound by the First Amendment, various theories justify the application of constitutional and statutory transparency rights.

A. The Company Town Theory

The Supreme Court has previously created multiple tests for determining whether extending First Amendment obligations to corporate actors is justified, each aiming to determine whether the corporate actor’s behavior amounts to state action. The four main tests are: (1) the public function test ; (2) the joint action test ; (3) the state compulsion test ; and (4) the governmental nexus test. Ultimately, these tests are all fact-bound inquiries that often conclude that if it walks like a duck and quacks like a duck, it is a duck; the more a corporate action appears to be a state action, the more likely it is one. Ultimately, these tests affirm the same thing, that there is some measure of corporate entanglement that creates a need for democratic accountability. “A basic rule, for example, is that a street or a park is a quintessential forum for the exercise of First Amendment rights,” and so the fact that a corporate entity holds title to that space, historically reserved for government, does not diminish its character as a public domain. Applying that rule, the Court has held that a private park operator is subject to the Fourteenth Amendment because a park is municipal in character. But the state action doctrine not only applies to “traditional public forums,” such as streets or parks ; the Court has also extended it to “designated public forums,” or nontraditional spaces that open “channel[s] of communication” and that “share essential attributes of a traditional public forum.”

Many social media companies arguably satisfy multiple of these tests, serving as the modern public parks and public forums in addition to providing multiple other traditional public functions. While no court has ever explicitly held that a social media company is a public actor, the Supreme Court has suggested a willingness to consider this as a viable theory. In Packingham v. North Carolina, a case striking down a North Carolina statute prohibiting registered sex offenders from accessing social networking websites, the Court in dicta equated social media sites to modern-day parks and streets, signaling that these companies take on responsibilities when they host a space of public activity. In that vein, some lower courts have held that social media pages have qualified as “public forums” where government exercises particular control over certain accounts. Today, when 70 percent of Americans access social media sites on a daily basis to engage in behaviors that would have traditionally occurred on our sidewalks, parks, and street corners, it is fair to say that these sites are participating in state action.

This argument is only further compounded by the recent development of many such titans functioning as more than just public forums, expanding into real-life “company towns.” Facebook, Amazon, and Google have all begun building housing, grocery stores, retail districts, and hotels in their respective Silicon Valley cities of Palo Alto, San Jose, and Mountain View as well as other cities across the country. “By 2021, Facebook is scheduled to complete Willow Park, a corporate campus with 1,500 housing units, retail, a hotel, and grassy plazas in Menlo Park, California.” At the same time, these companies are providing multiple government functions on the national level, including defense tools, banking features, and speech forums as well as modes of transit, as previously mentioned. Writing on a blog, Professor Ruthann Robson has recently argued for this position, citing Marsh v. Alabama, the landmark Supreme Court case that found that private companies that provide multiple government functions are bound by First Amendment obligations. Robson’s proposal for “arguments extending the Marsh company-town holding” is increasingly apropos as media companies “perform[] the full spectrum of municipal powers.”

Lastly, expanding the state action doctrine to these companies would align with the First Amendment’s right of the listeners. As Professor Helen Norton has written, the “law sometimes . . . puts listeners’ interests first in settings where those listeners have less information or power than speakers.” Often described as the listener’s right to receive information, this doctrine is usually raised to support the right of access to media outlets in cases involving newsworthy information. The fundamental idea in such instances is that when space is limited, access becomes paramount. As one court stated: “[W]hat exists of the right of access if it extends only to those who can squeeze through the door?” As scholar RonNell Anderson Jones has written “[a] recognition of [the listener approach] drives the growing movement to consider how listener rights might have distinct legal force.” It permits the government to regulate the speech of comparatively knowledgeable or powerful speakers when that expression frustrates their listeners. Under this approach, certain situations would justify favoring listeners, such as social media users, over speakers, such as social media companies, that engage in noisy and coercive algorithmic speech. 

B. Beyond Providing Fora: Controlling the New Corporate Towns’ Governors

The structural directive of the First Amendment also supports expanding the right of access to social media companies. As Justice William Brennan explained, “the First Amendment . . . has a structural role to play in securing and fostering our republican system of self-government.” “[S]tructuralist” strategies “limit the underlying powers and capacities” of powerful institutions in contrast to “prophylactic rules” that depend on “fine-tuning expert management.” Expanding the right of access is seemingly a banal solution to those who challenge the efficacy of transparency rights, but in terms of shifting power structures it is a radical move. Instead of behaving with deference to social media companies, this framework introduces disruption which is necessary for accountability and may at least slightly pull back the tide of the Lochner-ization of the First Amendment by redistributing the economic power of knowledge back to the public. In particular, the right of access, if expanded, could bring light to at least three areas of public life where technology companies enshroud information with secrecy anathema to our democractic values.

1. Access to Technology Records Involving Criminal Justice System

Access to information within the criminal justice system has always been viewed “at the core of First Amendment.” For this reason courts have gradually expanded the right beyond criminal trials, to voir dire proceedings, suppression hearings, and due process and entrapment hearings. Today, technology companies play instrumental roles in our criminal justice system, assisting the police, prosecutors, attorneys, and even judges with making crucial decisions that affect a person’s freedom. Judges often directly rely on risk assessment algorithms and other technological tools to decide crucial questions of criminal justice in pretrial detention, bail, sentencing, and parole. While companies making these technological tools claim they facilitate more neutral decision-making, these tools have been shown to be susceptible to various forms of bias. For instance, in 2016, ProPublica released an investigation showing the racial bias within COMPAS, an algorithm used by law enforcement agencies to assess risk. According to ProPublica, African-American defendants were incorrectly flagged as future criminals twice as frequently as non-African-American ones. In circumstances like these, technological tools integrally impact choices in the criminal justice system traditionally left to government actors, leading to the obvious conclusion that courts should expand the right of access to this information. While efforts to access information held by software developers are often met with claims that the information is proprietary or constitutes a protected trade secret, courts have historically countered those claims when they appear spurious.

2. Access to Technology Records Involving Civil Rights and Civil Liberties

Beyond law enforcement, algorithms and other technological tools have been used in decisions over whether to grant or deny a variety of other government benefits and civil rights and civil liberties. For instance, algorithms have been used in rationing healthcare. In one case, a court found that a state’s allotment of Medicaid – based on an algorithm – was so unreliable that it “arbitrarily deprive[d] participants of their property rights and hence violate[d] due process.” Similarly, public school teachers have been evaluated through privately developed software. There, the court wrote, “teachers have no meaningful way to ensure correct calculation of their [evaluation] scores, and as a result are unfairly subject to mistaken deprivation of constitutionally protected property interests in their jobs.” In cases involving due process as well as other rights and liberties, it is uniquely important that the right of access apply to ensure protections such as impartial adjudication as well as judicial review.

3. Access to Technology Records Mirroring Administrative or Judicial Bodies

Similarly, the right of access should also be expanded where technology companies create institutions that reproduce judiciary functions. Traditionally, the right of access has applied to a bounty of legal proceedings, including suppression hearings, bail hearings, and sentencing hearings as well as a plethora of court records such as indictments, motion documents, and criminal cases, among many others. Moreover, this right was eventually expanded to other court-like bodies outside the judicial system, such as the executive branch’s judicial review board proceedings, federal administrative fact-finding hearings, state legislative meetings, city council meetings, and governor's executive travel records. This past year, Facebook announced a “Global Oversight Board,” a Supreme Court–like body that will determine what speech on its platform may be censored. Expanding the right of access to the records and trial-like hearings of this type of judicial body determining core speech rights would appropriately fall in line with the structural demands of the First Amendment.

IV. Conclusion

As companies increase their control over our lives and our society, expanding the right of access is increasingly important. A constitutional guarantee of access to certain records and information held by technology companies would mark a significant step forward in checking the increasing power of these companies over people’s lives and over our democracy as a whole. While the passage of new FOIA-like legislation would also be appropriate, particularly to address the growing number of cases where the government tries to use Exemption 4 (trade secrets or confidential commercial information) to avoid disclosure requirements, this article has aimed to further the debate over how to transform our “black-box society” back into a democratic one. By employing a broader application of the right to access to technology companies we can assure more democratic stability in an institutional infrastructure of increasing asymmetry.