Abstract

The ChatGPT chatbot has not just caught the public imagination; it is also amplifying concern across industry, academia, and government policymakers interested in the regulation of Artificial Intelligence (AI) about how to understand the risks and threats associated with AI applications. Following the release of ChatGPT, some EU regulators proposed changes to the EU AI Act to classify AI systems like ChatGPT that generate complex texts without any human oversight as “high-risk” AI systems that would fall under the law’s requirements. That classification was a controversial one, with other regulators arguing that technologies like ChatGPT, which merely generate text, are “not risky at all.” This controversy risks disrupting coherent discussion and progress toward formulating sound AI regulations for Large Language Models (LLMs), AI, or ICTs more generally. It remains unclear where ChatGPT fits within AI and where AI fits within the larger context of digital policy and the regulation of ICTs in spite of nascent efforts by OECD.AI and the EU.

This paper aims to address two research questions around AI policy: (1) How are LLMs like ChatGPT shifting the policy discussions around AI regulations? (2) What lessons can regulators learn from the EU’s General Data Protection Regulation (GDPR) and other data protection policymaking efforts that can be applied to AI policymaking?

The first part of the paper addresses the question of how ChatGPT and other LLMs have changed the policy discourse in the EU and other regions around regulating AI and what the broader implications for these shifts may be for AI regulation more widely. This section reviews the existing proposal for an EU AI Act and its accompanying classification of high-risk AI systems, considers the changes prompted by the release of ChatGPT and examines how LLMs appear to have altered policymakers’ conceptions of the risks presented by AI. Finally, we present a framework for understanding how the security and safety risks posed by LLMs fit within the larger context of risks presented by AI and current efforts to formulate a regulatory framework for AI.

The second part of the paper considers the similarities and differences between the proposed AI Act and GDPR in terms of (1) organizations being regulated, or scope, (2) reliance on organizations’ self-assessment of potential risks, or degree of self-regulation, (3) penalties, and (4) technical knowledge required for effective enforcement, or complexity. For each of these areas, we consider how regulators scoped or implemented GDPR to make it manageable, enforceable, meaningful, and consistent across a wide range of organizations handling many different kinds of data as well as the extent to which they were successful in doing so. We then examine different ways in which those same approaches may or may not be applicable to the AI Act and the ways in which AI may prove more difficult to regulate than issues of data protection and privacy covered by GDPR. We also look at the ways in which AI may make it more difficult to enforce and comply with GDPR since the continued evolution of AI technologies may create cybersecurity tools and threats that will impact the efficacy of GDPR and privacy policies. This section argues that the extent to which the proposed AI Act relies on self-regulation and the technical complexity of enforcement are likely to pose significant challenges to enforcement based on the implementation of the most technologically and self-regulation-focused elements of GDPR.

Abstract

We thank the Federal Communications Commission for this opportunity to comment on the rules proposed in the above-captioned Notice of Proposed Rulemaking.[1] We are both legal academics with long-standing interest in the topics addressed by these proposed rules. Over the past 15 years, we have each published numerous articles, books, and other scholarly works on network neutrality and related topics, submitted comments in relevant Commission proceedings and briefs in related judicial proceedings, and been active participants in academic and public discussion of these topics.[2]

The views presented below are ours alone and should not be attributed to our employer or to the Center for Technology, Innovation, and Competition. Neither of us has received any compensation for these comments, nor has either of us been retained by any party with a financial interest in these proceedings.

The essence of the rules – their purpose and flaws – are effectively captured in the first paragraphs of the NPRM:

1. Today we propose to reestablish the Federal Communications Commission’s (Commission) authority over broadband Internet access service by classifying it as a telecommunications service under Title II of the Communications Act of 1934, as amended (Act). While Internet access has long been important to daily life, the COVID-19 pandemic and the rapid shift of work, education, and health care online demonstrated how essential broadband Internet connections are for consumers’ participation in our society and economy. Congress responded by investing tens of billions of dollars into building out broadband Internet networks and making access more affordable and equitable, culminating in the generational investment of $65 billion in the Infrastructure Investment and Jobs Act.
2. But even as our society has reconfigured itself to do so much online, our institutions have fallen behind. There is currently no expert agency ensuring that the Internet is fast, open, and fair. Since the birth of the modern Internet in the 1990s, the Commission had played that role, but the Commission abdicated that responsibility in 2018, just as the Internet was becoming more vital than ever.

While the NPRM would establish various specific rules (e.g., the proposed conduct rules and transparency rule) and puts forth various justifications for them, the primary and most significant goal of the proposal is to classify Broadband Internet Access Service (BIAS) as a Title II telecommunications service. That is literally the first sentence of the NPRM.

This purpose is also the most objectionable and least necessary aspect of the proposed rules. Also as noted in the first paragraph, and discussed in the NPRM, the COVID-19 pandemic made more clear than ever the importance of the Internet to modern life. But unrecognized by the NPRM, this period also demonstrated the extent to which the pervasive regulation contemplated by Title II is unnecessary for the Internet to satisfy this important role. While stressed, and while benefitting from support from programs like the Emergency Broadband Benefit,[3] American Rescue Plan Act,[4] and Affordable Connectivity Program,[5] the Internet proved its importance throughout the pandemic largely by living up to the tasks to which it was unexpectedly put. A lack of Open Internet regulations did not prevent it from doing so.

The first paragraph also recognizes Congress’s investment of $65 billion to support broadband deployment in response to the pandemic.[6] In so doing, Congress tasked an agency other than the Commission, the National Telecommunications and Information Agency (NTIA), with primary responsibility for overseeing use of this funding through the Broadband Equity, Access, and Deployment (BEAD) program.[7] And Congress did not include any network neutrality or open Internet provisions as requirements for allocation of this funding. This calls into substantial question whether Congress views the Commission as the regulatory agency with regulatory authority as relates to the Internet and, especially, whether Congress views rules such as proposed in the NPRM as necessary.

Even to the extent that it is true that the Commission is a relevant expert agency, the assertions in the second paragraph of the NPRM are false. The Federal Trade Commission (FTC) has expertise in antitrust and consumer protection. To the extent that Internet Service Providers (ISP) commit to providing fast, open, and fair service to their users—promises routinely made—there is, in fact, an agency with relevant expertise to ensure those commitments are kept. And to the extent that an ISP does not make such commitments, the First Amendment limits the FCC’s ability to impose them upon the ISP through the proposed rules. And to the extent that the FCC has played that role “since the birth of the modern Internet in the 1990s,” Title II classification has only been the basis for that role for the brief period between adoption of the 2015 Open Internet Order and its recission in the 2018 Restoring Internet Freedom Order.

The Commission is needlessly steering into political controversy through Title II reclassification. The D.C. Circuit’s 2014 opinion in Verizon makes clear that implementation of the substantive rules proposed in the NPRM could be accomplished using the Commission’s Title I authority.[8] There would likely be widespread support for (or at least acceptance of) rules adopted on this basis, both political and from industry.

This concern is heightened by the current judicial landscape. The Supreme Court’s evolving Major Questions Doctrine jurisprudence casts significant doubt on the Commission’s ability to assert long-disclaimed pervasive regulatory authority over Internet services—services that the Commission has repeatedly referred to as of substantial economic and political significance.[9] This is heightened by the changing technology of Internet access, which calls into doubt the ongoing relevance of the analysis in Brand X[10], the most important precedent supporting Title II classification of BIAS.[11] And the Supreme Court is hearing this term cases that bear directly on First Amendment central to the proposed rules.[12] The proposed rules significantly implicate speech regulation and are therefore highly sensitive to changing First Amendment jurisprudence. Proposing rules prior to resolution of these ongoing cases creates a needles risk of uncertainty over the coming years—a risk that directly contradicts the importance that the Commission ascribes to the proposed rules.

The remainder of these comments further develop these concerns in four sections. Part I discusses how technological developments over the past twenty years raise doubt about the continued application of the conclusions in Brand X. Part II looks at the current state of the broadband Internet services industry, focusing on the BEAD Program. Part III looks at First Amendment considerations relating to the NPRM. And Part IV considers the Supreme Court’s recent Major Questions Doctrine jurisprudence.

I. Changes in BIAS Architecture Cast Doubt on Continued Relevance of Brand X

The most important precedent governing classification of BIAS as a Title I information service or Title II telecommunication service is the Supreme Court’s opinion in Brand X. The question in Brand X was, in a sense, the opposite of that presented by the NPRM.

Both involve the classification of broadband Internet service as either an “information service” (regulated under Title I of the Communications Act) or a “telecommunications service” (regulated under Title II of the Communications Act). These terms are defined in 47 U.S.C. 153:

(20) INFORMATION SERVICE.—The term “information service” means the offering of a capability for generating, acquiring, storing, transforming, processing, retrieving, utilizing, or making available information via telecommunications, and includes electronic publishing, but does not include any use of any such capability for the management, control, or operation of a telecommunications system or the management of a telecommunications service.

(43) TELECOMMUNICATIONS.—The term “telecommunications” means the transmission, between or among points specified by the user, of information of the user’s choosing, without change in the form or content of the information as sent and received.

(46) TELECOMMUNICATIONS SERVICE.—The term “telecommunications service” means the offering of telecommunications for a fee directly to the public, or to such classes of users as to be effectively available directly to the public, regardless of the facilities used.

In Brand X, the Court recognized that some portion of the Internet service offered by a cable ISP is a telecommunications service and that some other portion is an information service.[13] The question the Court faced was whether those were so tightly integrated that the entire offering could be classified as a single information service offering, as opposed to separate and distinct telecommunications and information service offerings.[14]

In the proposed rules, the Commission proposes to do the exact opposite: they would classify an offering that includes some amount of information services alongside a telecommunications service as an integrated telecommunications service. While it seems natural to characterize these as obverse sides of a coin, the statutory reality does not allow this approach. The definition of an “information service” is the offering of various capabilities “via telecommunications”—information services are, in other words, necessarily integrated with a telecommunications service. The definition of “telecommunications service,” on the other hand, includes exclusively the provision of telecommunications. Any blending of these services can only be an information service.

If the Commission is to classify BIAS as a Title II telecommunications service, such classification must apply solely to the portion of an ISP’s service that is not an information service. If, in other words, Domain Name System (DNS) services are information services, offering consumers a service that includes DNS access must necessarily fall outside of the scope of BIAS offerings classified as Title II services. The same applies for any other services offered by an ISP that are necessarily information services.

As noted in the NPRM, a similar argument was rejected by the D.C. Circuit Court of Appeals in its affirmation of the 2015 Open Internet Order’s classification of BIAS as a Title II service.[15] There, a two-judge majority accepted the Commission’s characterization of DNS service as relating to the “management, control, or operation of a telecommunications system or the management of a telecommunications service,” features which remove the service from the statutory definition of an information service.[16] Dissenting, Judge Williams would have rejected that argument as arbitrary and capricious.[17]

Even accepting the Commission’s argument, this approach turns the question into a technical one. Judge Williams notes that with this approach “the Commission set for itself a highly technical task of classification”[18]—an echo of the Supreme Court’s statement in Brand X that the question of whether transmission services and DNS service are functionally integrated “turns not on the language of the Act, but on the factual particulars of how Internet technology works and how it is provided.”[19]

The question of whether DNS service is a management, control, or operation function was briefly discussed in Brand X, as well—though it was not properly before or decided by the Court. Rather, it was raised by Justice Scalia in his dissent, to which the majority responded in its footnote 3. In his dissent, Justice Scalia characterizes DNS service as “scarcely more than routing information, which is expressly excluded from the definition of ‘information service.’”[20] In its footnote, the majority notes that “routing information” is not, in fact, among the functions excluded from the definition of an information service.[21] Of course, ISPs do offer routing services. They do not merely transmit information of the user’s choosing between or among points specified by the user without change in the form or content sent or received—they very often select for the user which endpoints to which the user’s information will be sent or from which it will be retrieved and the routes by which that information will be sent and received. And this is very often accomplished using DNS configurations.[22]

This is significantly more true today than it was at the time of the DC Circuit’s review of the 2015 Open Internet Order, and even more so true than it was at the time of the Brand X case. The highly technical nature of the Internet, including the features provided by some, but not all, ISPs, has changed dramatically since the 2015 Open Internet Order—and it continues to change. Some ISPs are working to deploy technologies like the Internet Engineering Task Force’s proposed Non-Queue-Building Per-Hop Behavior standard to improve DNS performance.[23] Some ISPs are working to support development of new active queue management technologies, such as the Internet Engineering Task Force’s proposed Low Latency, Low Loss, Scalable Throughput standard, which, again, have the potential to dramatically improve performance of Internet services.[24] Larger ISPs can implement dynamic routing across their networks to improve performance.[25] These are all service that some, but not all, ISPs are likely to implement to offer than enhance users’ ability to acquire, retrieve, use, or make available information—and in their improved forms none is necessary to the management or operation of the underlying telecommunications service.

Similarly, the necessity of DNS services provided by ISPs has only decreased in recent years. Today there are several free, open DNS services, including services provided by Google and Cloudflare. Statistics suggest that at least 20% of Internet users use these services instead of DNS services provided by their ISP.[26] There are parts of the world in which such services account for more than 50% of the market, demonstrating the extent to which such services are not necessary to the management or operation of the underlying telecommunications system.[27] Private VPN services are widely advertised and obviate the need for an ISP’s DNS services. Web browsers such as Chrome and Firefox include support for DNS over HTTPS, including their own DNS server settings—indeed, by default Firefox bypasses the ISPs’ DNS servers entirely.[28]

Together, these demonstrate the extent to which even a mundane-seeming service like DNS is more than a mere network management feature. While Justice Scalia’s argument in Brand X that DNS is not an information service might have been reasonable at the time, with each passing year that argument becomes weaker. Today, DNS is a standalone service, undergoing active research and development, offered in a competitive marketplace. Consumers can, and do, bypass their ISPs’ DNS services; standalone products like web browsers and VPN services can, and do, bypass their users’ ISPs. And ISPs invest in improving the quality of their own DNS services not because it improves their management of the service—marginal increases in DNS performance offer zero benefit to the management, control, or operation of a telecommunications system—but because it enhances the users’ ability to generate, acquire, store, transform, process, retrieve, utilize, or make available information via telecommunications.

II. Ongoing Development of the Broadband Industry, Notably Including Congressional Broadband Programs, Cast Doubt on the Proposed Rules

The NPRM expresses concern that ISPs “have the incentive and ability to engage in practices that pose a threat to Internet openness” and seeks comment on the state of competition in the market.[29] Contrary to the NPRM’s concerns, the state of competition in the BIAS market is robust and continues to increase. While the NPRM cites to 2021 data that that approximately 36 percent of households do not have access to two or more providers offering 100/20 Mbps wireline Internet service, that same data shows that more than 86 percent of households do have competitive options at the 25/3 Mbps level—even where a customer may only have a single ISP offering 100/20 Mbps speeds, almost all of those customers have at least one other option that creates competitive pressure.[30]

But the NPRM’s cited data is subject to greater criticism on other fronts. First, it cites to service availability data from 2021—a period during which ISPs were rapidly expanding their capacity to accommodate pandemic-era demands. The rate of this expansion was supplement by tens of billions of dollars of funding was made available through the American Rescue Plan Act (ARPA) to support broadband infrastructure investment, none of which is included in the cited data.[31] This data also doesn’t consider satellite options—Space X’s Starlink is now widely available throughout the United States[32] and Amazon’s Kupier project is beginning to put satellites into orbit.[33] And it excludes consideration of fixed wireless services, including both those commonly offered by WISPs and new 5G fixed wireless services offered by traditional wireless companies like T-Mobile and Verizon. The exclusion of fixed wireless services is notable given the pace at which consumers are embracing the technology as an alternative to traditional wireline service options.[34]

And, of course, the numbers cited by the NPRM exclude future broadband availability that will be facilitated by the Broadband Equity Access and Deployment (BEAD) program, which will increase broadband availability in currently un- and underserved areas.

The BEAD program draws attention to another concern about the proposed rules: over the past several years Congress has enacted numerous laws that support broadband investment but has not included net neutrality or open internet considerations in any of them. As will be discussed in the final section of these comments, the ultimate question for any federal agency action is whether it is authorized by Congress. The concept of net neutrality has been debated in Congress for more than a decade—it not conceivable that Congress was not aware of concerns such as those animating this NPRM as it developed the BEAD program. An inference can therefore be drawn from the fact that Congress, in allocating tens of billions of dollars in support for broadband investment, did not think it necessary to include provisions relating to network neutrality, even though doing so would settle a debate that has spanned four presidential administrations and produced substantial uncertainty within industry.

Indeed, a related inference might be drawn from Congress’s decision to entrust the National Telecommunications and Infrastructure Agency to oversee the BEAD program’s more-than $42 billion budget.[35] The BEAD program has many similarities to the Commission’s longstanding universal service programs, including raising many issues about which one would ordinarily assume the Commission as—and is viewed by Congress as having—substantial expertise. That Congress turned to another agency for the implementation of the largest broadband infrastructure program in the nation’s history should provide guidance on how broadly to interpret the Commission’s authority under the Communications Act to implement related program such as those proposed in the NPRM.[36]

III. First Amendment Considerations Cast Doubt on the Proposed Rules

The proposed rules raise at least three sets of concerns under the First Amendment. First, whether imposition of common carriage obligations is problematic under First Amendment principles. Second, whether the specific proposed rules are problematic under First Amendment principles, even absent reclassification. And third, the relevance of cases currently pending before the Supreme Court to the proposed rules.

The last of these concerns can be addressed with brevity: NetChoice v. Paxton is currently pending before the Supreme Court, with an opinion expected this term.[37] This case asks the Court to consider whether laws that limit social media platforms’ content-moderation practices, imposing common-carriage-like obligations upon them, comply with the First Amendment. The laws that the Court is reviewing bear similarity to the rules proposed by the Commission and there is a high likelihood that the Court’s decision with have significant bearing on these proposed rules. It would be imprudent to adopt any rules on this topic until the Supreme Court issues its opinion in NetChoice.

The NetChoice case also illustrates the next point to consider: whether imposition of common carriage obligations is problematic under the First Amendment. There is a long history of efforts to impose such obligations on communications platforms, from the social media at issue in NetChoice, to radio and television broadcasters, news papers, cable networks, and telephone networks.[38] But it turns out that the idea of “imposing common carriage obligations” gets the question backwards. The most universally accepted definition of common carriage turns on whether the firm eschews exercising editorial discretion over the content it carries and instead holds itself out as serving all members of the public without engaging in individualized bargaining.[39] In other words, the question is not whether government can impose common carriage obligations—it is whether a firm conducts its business in the style of a common carrier. Doing so can create an obligation to honor the obligations of common carriage (which may come with regulatory benefits). But a firm can avoid those obligations by engaging in individualized bargaining. This criterion constitutes the central consideration in all leading discussions of common carriage.[40]

This leads to the curious conclusion that the Commission’s proposal to classify BIAS as a Title II telecommunications service—a common carriage service—would only have effect to the extent that a BIAS provider elects to hold itself out as offering a common carriage service. This conclusion is reflected in the exchange that took place during the D.C. Circuit’s decision not to rehear the decision upholding the 2015 Open Internet Order en banc. When then-Judge Kavanaugh objected that classifying ISPs as common carriers impermissibly abridged their editorial discretion,[41] the authors of the majority opinion countered, explaining that “[w]hen a broadband provider holds itself out as giving customers neutral, indiscriminate access to web content of their own choosing, the First Amendment poses no obstacle to holding the provider to its representation.”[42] The 2015 Open Internet Order did not implicate the First Amendment because it only purported to regulate services over which providers exercised no editorial discretion.[43] This discussion implicitly recognized that speech over which providers exercise editorial control is protected by the First Amendment. If that were not the case, the fact that the 2015 Open Internet Order affected only speech over which providers exercised no editorial discretion would have been completely unresponsive to the concerns raised by then-Judge Kavanaugh. As described by the majority, this is a “if you say it, do it” theory, nothing more.[44]

Indeed, the initial Circuit Court opinion reached a similar conclusion—and noted the 2015 Open Internet Order recognized as much as well:

If a broadband provider nonetheless were to choose to exercise editorial discretion—for instance, by picking a limited set of websites to carry and offering that service as a curated internet experience—it might then qualify as a First Amendment speaker. But the Order itself excludes such providers from the rules. . . . Providers that may opt to exercise editorial discretion—for instance, by offering access only to a limited segment of websites specifically catered to certain content—would not offer a standardized service that can reach “substantially all” endpoints. The rules therefore would not apply to such providers, as the FCC has affirmed.[45]

The perverse incentive this creates bears emphasis. It is clear from the history of the Commission’s Open Internet proceedings that BIAS providers do not want to be subject to Title II regulation. These providers might today hold themselves out as providing services that might be considered by consumers to be common carriage services. If these providers want to avoid the obligations of Title II regulation, they would be able to do so by reducing the scope of their services to the point that they cannot be seen as holding themselves out as common carriers. The First Amendment affords no path for the Commission to compel ISPs to do otherwise.

A question still remains whether the specific rules proposed in the NPRM are problematic under First Amendment principles, even absent reclassification. In Turner, the Supreme Court upheld “must-carry” rules for cable networks, which are not common carriers, even where those networks have some claim to editorial discretion.[46] The result in Turner, however, turned on the “gatekeeper” or “bottleneck” control resulting from “the fact that there could only be one cable connection to any home” that places the cable operator in a position to block any other content providers from gaining access to subscribers.[47] In so holding, the Court emphasized the physical (rather than economic) nature of this consideration by contrasting cable with newspapers, which, “no matter how secure its local monopoly, does not possess the power to obstruct readers’ access to other competing publications.”[48] In an era of multimodal communications, where households routinely have connections to telephone, cable, and fiber optic networks, as well is fixed wireless, mobile wireless, and satellite connectivity options, makes clear the inapplicability of this rationale to platforms that lack control over an exclusive physical connection precludes applying this rationale to ISPs.

IV. The Major Questions Doctrine Casts Doubt on the Proposed Rules

The NPRM asks dozens of questions over several paragraphs relating to the Major Questions Doctrine.[49] First formally recognized by the Supreme Court in 2022, this doctrine explains that the Court “expect[s] Congress to speak clearly if it wishes to assign to an agency decisions of vast ‘economic and political significance.’”[50] Although only first recognized by the Court in 2022, the Major Questions Doctrine has developed over a span of many decades, through both Supreme Court cases and those decided by lower courts.[51] Importantly, nothing in these cases suggests that an agency’s recognition that its decisions may raise major questions renders those decisions any less major. Rather, the fact that an agency feels it is necessary to ask whether its decisions raise major questions suggests that those questions may well be major. This alone should give the agency pause about taking such decisions—especially in an era of intense judicial scrutiny of agency action it would a curious decision for any agency instead seek to structure its decisions so as to avoid the appearance of their having vast economic or political significance.

The vast significance of the proposed rules cannot be overstated—though the NPRM seems notably to attempt to understate it. Discussing broadband in 2015, former Chair Tom Wheeler described the Internet as “the most powerful network in the history of mankind.”[52] This was echoed in the 2015 Open Internet Order. The first sentence of the 2015 Order asserted that “[t]he open Internet drives the American economy and serves, every day, as a critical tool for America’s citizens to conduct commerce, communicate, educate, entertain, and engage in the world around them.”[53] Similarly, the NPRM for that Order began with: “The Internet is America’s most important platform for economic growth, innovation, competition, [ and] free expression . . . . [It] has been, and remains to date, the preeminent 21st century engine for innovation and the economic and social benefits that follow.”[54]

And, as made clear in the current NPRM, this importance has only been amplified since the beginning of the COVID-19 pandemic: “In the time since the RIF Order, propelled by the COVID-19 pandemic, BIAS has become even more essential to consumers for work, health, education, community, and everyday life.”[55] As stated by Chair Rosenworcel, “broadband is no longer nice-to-have; it’s need-to-have for everyone, everywhere. Broadband is an essential service”[56] Commissioner Starks similarly states that “there is simply no way to overstate broadband’s impact on the lives of individual Americans.”[57] And Commissioner Gomez states “El acceso a internet de banda ancha no solo es una herramienta vital para la educación, la atención de salud y para comunicarnos con nuestros seres queridos. También es un conducto de crítica importancia, esencial para la vida moderna.”[58]

In the NPRM, the Commission points to the D.C. Circuit’s opinion in the previous Open Internet Order case, as well as to Brand X itself, to suggest that the Brand X opinion settles the question of whether the Commission has authority to adopt these rules.[59] As an initial matter, that question simply was not before the Court in Brand X. In Brand X, the Court considered only whether the FCC could decide the classification under the Chevron after a lower court had adopted a conflicting interpretation of an ambiguous statute.[60] The question of whether Commission had authority to regulate cable broadband Internet service as a Title II telecommunications service was not a contested issue—only whether it could instead interpret ambiguity in the Communications Act to instead regulate it as a Title I information service. Similarly, the Major Questions Doctrine had not been expressly recognized by the Supreme Court at the time of the D.C. Circuit’s 2016 opinion. While it was “in the air,” it was a controversial doctrine—one that the Supreme Court had yet to expressly embrace. But the Court has done so now.

In the D.C. Circuit’s denial of en banc review, the judges Srinivasan and Tatel (the authors of the majority opinion under review) offers a more fulsome consideration of the major questions issue[61]—prompted by the dissenting opinion of then-judge Kavanaugh.[62] But in light of the Supreme Court’s own more fulsome embrace of the doctrine since that opinion, this analysis is unconvincing. In addition to over-relying on Brand X as having decided the matter, the majority makes another conceptually uncertain assumption: even if the Commission could have regulated Internet services at the time of Brand X, that doesn’t necessarily mean that the Commission has such authority today. The economic and political significance of those services is vastly different today than it was then. The Commission disclaimed such authority for more than 15 years, during which time the social, economic, technical, and political understandings of the Internet changed fundamentally from what they were at the time of Brand X. As discussed in the following paragraphs, this consideration is relevant to the Major Questions Doctrine analysis but goes unrecognized by the D.C. Circuit opinion.

While Brand X does not support the contention that classifying BIAS as a Title II does not present major question, it does briefly draw attention to an important argument that it does present a major question: respondents in Brand X raised the concern that “the Commission’s construction [of cable Internet as a Title I information service] is unreasonable because it allows any communications provider to ‘evade’ common-carrier regulation by the expedient of bundling information service with telecommunications.”[63] The Court rejected the premise of this argument, so did not decide whether it would, in fact, be an unreasonable construction.[64] But the same issue runs through the proposed rules: that BIAS providers need not hold out their offerings as common carriage services, so could “evade” Title II regulation through a simple expedient. Unlike in Brand X, where the Court rejected the premise that such evasion was possible, here the D.C. Circuit has found that the First Amendment compels such a result and the Commission has conceded that the rules necessarily allow it.[65]

This is a facially absurd result that demonstrates the fundamental mismatch between Title II regulation at BIAS service. Title II is a pervasive regulatory regime designed to regulate one of the foundational, utility-style, natural monopoly, industries of the 20^th century. It was designed for a technology to which editorial discretion was largely viewed as inapplicable.[66] And it was part of a regulatory quid pro quo that gave telephone carriers a privileged regulatory position in exchange for offering service on a common-carriage basis. As we approached the 21^st century, Title II was substantially amended through the 1996 Telecommunications Act to be a pervasively deregulatory statute that would phase out regulate in favor of competition and did so largely in response to the development of networks that offered advanced communications capabilities that were increasingly removed from the transparent, point-to-point, communications model of the 20^th century telephone network.[67]

It is little surprise that the Commission’s efforts today to reregulate modern communications networks under an expressly deregulatory act runs into problems such as the possibility for evasion. Similarly, it is unsurprising that it required tailoring of the Title II regulatory regime through extensive use of forbearance—another hallmark of a regulatory decision that presents major questions.[68]

The idea that Congress intends the Commission to be the primary regulator for BIAS services faces even greater headwind following the pandemic. Congress—which has longstanding awareness of the net neutrality debates—put in place significant communications-related regulations with the IIJA. This included the BEAD program, the Affordable Connectivity Program, and laying the groundwork for the Commission’s recently proposed Digital Discrimination rules.[69] This legislation, adopted during a period of single-party control of Congress and the White House, could have easily clarified the Commission’s regulatory authority over BIAS. Instead, it assigned primary authority for the federal government’s flagship broadband infrastructure program to another agency and assigned to the Commission new authority to adopt more limited broadband discrimination rules to ensure equitable access to that infrastructure.

These circumstances bear resemblance to the circumstances discussed by the Court in FDA v. Brown & Williamson Tobacco Corp, one of the precursor cases to the Major Questions Doctrine.[70] In Brown & Williamson, the Court held that the FDA did not have authority to regulate tobacco, a drug, under its statutory authority to regulate drugs. In reaching this conclusion, the Court recognized among other things that Congress knew the FDA did not have a clear claim to regulate tobacco and had adopted alternative regulatory regimes relating to the regulation of tobacco without giving the FDA regulatory authority over the matter.[71] It also recognized that “it is hardly conceivable that Congress—and in this setting, any Member of Congress—was not abundantly aware of what was going on.”[72] Here, as there, if Congress intended the FCC to exercise the vast authority that the Commission would claim in the NPRM, Congress was aware of the uncertainty over the Commission’s authority and could readily have clarified the matter.

Conclusion

The defining feature of the Safeguarding and Securing the Open Internet NPRM is its proposed reclassification of BIAS as a Title II telecommunication service. By choosing this approach, the Commission is needlessly steering into both legal uncertainty and political controversy. Reclassification will be challenged in court; these challenges have a high likelihood of success and, in the best case, will yield years of uncertainty for industry, consumers, and the Commission. Indeed, cases currently pending before the Supreme Court could force reconsideration of the NPRM before the proposed rules could even be finalized.

This is a curious approach to take, given that ISPs can, in effect, “opt out” of reclassification by offering an Internet experience that does not purport to be a common carriage service. This would be an unfortunate outcome that would lead to consumer confusion and possibly to degraded user experiences. It is a perverse approach to regulation, imposing burdensome rules that can be avoided by degrading the quality of service offered to consumers. And this demonstrates the extent to which the proposed reclassification is a perversion of the purposes of Title II—and to which it therefore is a decision that raises major questions.

And the approach is all the more curious given that there is little demonstrable need for the rules in the first place. But if the rules are to be adopted, the Commission could do so without reclassification. Were the Commission to follow the roadmap offered by the D.C. Circuit in Verizon, it is entirely likely that industry would acquiesce to the rules. That is the path the Commission should take, if it is to go down this path at all.

[1]    Safeguarding and Securing the Open Internet, WC Docket No. 23-230, Notice of Proposed Rulemaking, 88 Fed. Reg. 76,048 (Nov. 3, 2023) (NPRM).

[2]    Further detail on many of the topics discussed in these comments can be found in our prior publications. See, e.g., [[past comments, briefs, articles]].

[3]    Consolidated Appropriations Act, 2021, Pub. L. No. 116-260, § 904 (2020).

[4]    American Rescue Plan Act of 2021, Pub. L. No. 117-2, § 7402 (2021) (ARPA).

[5]    Infrastructure Investment and Jobs Act, Pub. L. No. 117-58, § 60502 (2021) (IIJA)

[6]    Id. § 60101, et seq.

[7]    Id. § 60102.

[8]    See Tom Wheeler, Finding the Best Path Forward to Protect the Open Internet (April 29, 2014), available at https://www.fcc.gov/news-events/blog/2014/04/29/finding-best-path-forward-protect-open-internet (“In its Verizon v. FCC decision the D.C. Circuit laid out a blueprint for how the FCC could use Section 706 of the Telecommunications Act of 1996 to create Open Internet rules that would stick. I have repeatedly stated that I viewed the court’s ruling as an invitation that I intended to accept.”).

[9]    See Part IV.

[10]   Nat’l Cable & Telecomms. Ass’n v. Brand X, 545 U.S. 967 (2005) (Brand X).

[11]   See Part I.

[12]   See Part III.

[13]   Brand X at 989 (“Cable companies in the broadband Internet service business ‘offer’ consumers an information service in the form of Internet access and they do so “via telecommunications.”).

[14]   Id. At 990 (“The question, then, is whether the transmission component of cable modem service is sufficiently integrated with the finished service to make it reasonable to describe the two as a single, integrated offering”). More precisely, the Court was considering whether this interpretation, made by the FCC, can be sustained under the Chevron doctrine. Id. at 989 (“This construction passes Chevron’s first step.”); id. at 997 (“We also conclude that the Commission’s construction was ‘a reasonable policy choice for the Commission to make’ at Chevron’s second step.”).

[15]   NPRM, paras. 11, 75–77 (citing U.S.Telecom Ass’n v. FCC, 825 F.3d 674(D.C. Cir. 2016) (USTA))

[16]   USTA at 705.

[17]   Id. at 766, n.8 (Williams, J, dissenting).

[18]   Id. at 748.

[19]   Brand X at 991.

[20]   Id. at 1012–13 (Scalia, J, dissenting).

[21]   Id. at 999, n.3.

[22]   See Reply Comments of Christopher S. Yoo, WC Docket No. 17-108.

[23]   A Non-Queue-Building Per-Hop Behavior (NQB PHB) for Differentiated Services, IETF Transport Area Working Group Draft (Oct. 24, 2022), available at https://www.ietf.org/archive/id/draft-ietf-tsvwg-nqb-14.html.

[24]   Low Latency, Low Loss, Scalable Throughput (L4S) Internet Service: Architecture, IETF Transport Area Working Group Draft  (July 27, 2022), available at https://www.ietf.org/archive/id/draft-ietf-tsvwg-l4s-arch-19.html. See also Mitchell Clark, The Quiet Plan to Make the Internet Feel Faster, The Verge (Dec. 9, 2023).

[25]   See, e.g., Joan Feigenbaum et al, A BGP-based Mechanism for Lowest-Cost Routing, 18 Distributed Computing 61 (2005).

[26]   Geoff Huston, Looking at Centrality in the DNS,APNIC Blog (Nov. 22, 2022), available at  https://blog.apnic.net/2022/11/22/looking-at-centrality-in-the-dns/ (“the use of ISP-provided recursive resolution occurs for between 65% to 80% of users . . . known open resolvers have a 20% market share”).

[27]   Id.

[28]   Firefox DNS-over-HTTPS, Mozilla Support, available at https://support.mozilla.org/en-US/kb/firefox-dns-over-https (“When DoH is enabled, Firefox by default directs DoH queries to DNS servers that are operated by a trusted partner . . . .”) (visited Dec. 13, 2023).

[29]   NPRM, paras 126, 128.

[30]   Importantly, and contrary to the longstanding policy approach of defining performance metrics that focus on increasingly higher bandwidth (Mbps targets), “It has been demonstrated that, once access network bit rates reach levels now common in the developed world, increasing link capacity offers diminishing returns if latency (delay) is not addressed.” Low Latency, Low Loss, Scalable Throughput (L4S) Internet Service: Architecture, supra note 24. Implementation of new technologies, such as L4S, could increase competition between existing networks than at lower cost than building out entirely new, increasingly higher-speed, infrastructure.

[31]   See Anna Read & Kelly Wert, How States Are Using Pandemic Relief Funds to Boost Broadband Access, Pew (Dec. 6, 2021), available at https://www.pewtrusts.org/en/research-and-analysis/articles/2021/12/06/how-states-are-using-pandemic-relief-funds-to-boost-broadband-access.

[32]   Starlink Availability Map, available at https://www.starlink.com/map.

[33]   See Joey Roulette, Amazon’s Prototype Kuiper Satellites Operating Successfully, Reuters (Nov. 16, 2023), available at https://www.reuters.com/technology/amazons-prototype-kuiper-satellites-operate-successfully-2023-11-16/.

[34]   See Mike Dano, T-Mobile Exceeds in Q3, Talks Broadband Strategy, Light Reading (Oct. 25, 2023), available at https://www.lightreading.com/fixed-wireless-access/t-mobile-exceeds-in-q3-talks-broadband-strategy (“T-Mobile reported a total of 557,000 new fixed wireless access (FWA) customers, a figure above most analyst expectations and slightly ahead of the company’s recent quarterly pace.”); Mike Dano, FWA Captures 90% of All New US Customers, Pleasing Around 90% of Them, Light Reading (March 6, 2023), available at https://www.lightreading.com/fixed-wireless-access/fwa-captures-90-of-all-new-us-customers-pleasing-around-90-of-them (“fixed wireless services accounted for 90% of all net broadband customer additions in the US during 2022 . . . . 90% rated their service as ‘good enough.’”).

[35]   IIJA § 60502(b)(2).

[36]   See infra, Part IV.

[37]   Moody v. NetChoice, LLC; NetChoice, LLC v. Moody; NetChoice, LLC v. Paxton, Nos. 22-277, 22-393 and 22-555 (U.S. Sup. Ct.).

[38]   See generally Christopher Yoo, Free Speech and the Myth of the Internet as an Unintermediated Experience, 78 Geo. Wash. L. rev. 697, Part II (2010) (surveying regulation of various media technologies).

[39]   See Biden v. Knight First Amendment Inst., 141 S. Ct. 1220, 1222 (2021) (Thomas, J., concurring).

[40]   See FCC v. Midwest Video Corp., 440 U.S. 689, 701 (1979); U.S. Telecom Ass’n v. FCC, 825 F.3d 674, 740 (D.C. Cir. 2016); Verizon v. FCC, 740 F.3d 623, 651 (D.C. Cir. 2014); Cellco P’ship v. FCC, 700 F.3d 534, 548 (D.C. Cir. 2012); NARUC II, 533 F.2d at 608; NARUC I, 525 F.2d at 641. Congress, courts, and agencies have applied the same formulation in a wide variety of contexts. See 15 U.S.C. § 375(3); 46 U.S.C. § 40102(7)(A); 40 C.F.R. § 202.10(b); Edwards v. Pac. Fruit Express Co., 390 U.S. 538, 540 (1968); Woolsey v. Nat’l Transp. Safety Bd., 993 F.2d 516 524 n.2. (5th Cir. 1993); Flytenow, Inc. v. FAA, 808 F.3d 882, 887–88 (D.C. Cir. 2015); Nichimen Co. v. M. V. Farland, 462 F.2d 319, 326 (2d Cir. 1972); Kelly v. Gen. Elec. Co., 110 F. Supp. 4, 6 (E.D. Pa.), aff’d, 204 F.2d 692 (3d Cir. 1953).

[41]   U.S. Telecom Ass’n, 855 F.3d at 484–89 (Kavanaugh, J., dissenting from the denial of rehearing en banc)

[42]   Id. at 392 (Srinivasan, J., joined by Tatel, J., concurring in the denial of the petition for rehearing en banc).

[43]   Id. at 388–89.

[44]   Id. at 392.

[45]   USTA at 743.

[46]   Turner Broad. Sys., Inc. v. FCC (Turner I), 512 U.S. 622, 656–57 (1994).

[47]   Id. at 656.

[48]   Id.

[49]   NPRM, paras 81–84.

[50]   West Virginia v. EPA, 142 S. Ct. 2587 (2022) (quoting Utility Air Regulatory Group v. EPA, 573 U.S. 302, 324 (2014)).

[51]   Id. at 2609 (“major questions doctrine label . . . took hold because it refers to an identifiable body of law that has developed over a series of significant cases all addressing a particular and recurring problem: agencies asserting highly consequential power beyond what Congress could reasonably be understood to have granted. Scholars and jurists have recognized the common threads between those decisions. So have we.” (citing cases dating to MCI Telecommunications Corp. v. American Telephone & Telegraph Co., 512 U.S. 218 (1994) (MCI)).

[52]   See Remarks of FCC Chairman Tom Wheeler, Silicon Flatirons Center (Feb. 9, 2015), available at http://transition.fcc.gov/Daily_Releases/Daily_Business/2015/db0209/DOC-331943A1.pdf; see also of same, available at https://www.youtube.com/watch?v=vHsHkKpxVkQ (in which Chairman Wheeler was more emphatic than in his prepared remarks); Brian Fung, FCC chairman warns: The GOP’s net neutrality bill could jeopardize broadband’s ‘vast future’, Washington Post (Jan. 29, 2015), available at http://www.washingtonpost.com/blogs/the-switch/wp/2015/01/29/fcc-chairman-warns-that-republican-bill-couldjeopardize-broadbands-vast-future/.

[53]   Protecting and Promoting the Open Internet, GN Docket No. 14-28, Report and Order on Remand, Declaratory Ruling, and Order, 80 Fed. Red. 19,737 (Apr. 15, 2015).

[54]   Protecting and Promoting the Open Internet, GN Docket No. 14-28, Notice of Proposed Rulemaking, 79 Fed. Reg. 37,448 (July 1, 2014).

[55]   NPRM, para 16.

[56]   NPRM, Statement of Chairwoman Jessica Rosenworcel.

[57]   NPRM, Statement of Commissioner Geoffrey Starks.

[58]   NPRM, Statement of Commissioner Anna M. Gomez.

[59]   NPRM, para 81 (“In the USTA decision, the D.C. Circuit reasoned that Brand X conclusively held that the Commission has the authority to determine the proper statutory classification of BIAS and that its determinations are entitled to deference, and so there is no need to consult the major questions doctrine here.”).

[60]   Brand X, at 974 (“We must decide whether [the FCC’s] conclusion [that cable companies that sell broadband Internet service do not provide ‘telecommunications service’ as the Communications Act defines that term] is a lawful construction of the Communications Act under Chevron U. S. A. Inc. v. Natural Resources Defense Council, Inc. and the Administrative Procedure Act. We hold that it is.”).

[61]   U.S. Telecom Ass’n, 855 F.3d at 385–88.

[62]   U.S. Telecom Ass’n, 855 F.3d at 422–26 (Kavanaugh, J., dissenting).

[63]   Brand X, at 997.

[64]   Id. (“We need not decide whether a construction that resulted in these consequences would be unreasonable because we do not believe that these results follow from the construction the Commission adopted.”).

[65]   See supra, Part III.

[66]   But see Christopher Yoo, Free Speech and the Myth of the Internet as an Unintermediated Experience, 78 Geo. Wash. L. rev. 697, 752–57 (2010) (discussing cases starting in the 1980s that began to recognize the applicability of editorial discretion concepts to telephone carriers).

[67]   See Telecommunications Act of 1996, Pub. L. No. 104-104 (1996) (“An act to promote competition and reduce regulation in order to secure lower prices and higher quality services for American telecommunications consumers and encourage the rapid deployment of new telecommunications technologies.”).

[68]   See U.S. Telecom Ass’n, 855 F.3d at 404–09 (Brown, J., dissenting from the denial of rehearing en banc) (citing MCI and Utility Air Regulatory Group v. EPA in discussing the problematic use of the Commission’s forbearance authority to “tailor” Title II to fit the needs of BIAS regulation).

[69]   See supra, notes 3–5.

[70]   FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120 (2000).

[71]   Id. at 155-156.

[72]   Id. at 156.

Abstract

Once considered the final frontier, outer space has become the modern day Yukon territory. A burgeoning commercial economy is reshaping the balance of powers and expanding the breadth of activities beyond our atmosphere. Outer space is no longer the exclusive province of a select number of nation states engaged in geopolitical competition. A robust private sector has begun to stake its claim, ushering in a fundamentally different incentive environment that answers to shareholders and venture financers. As a consequence, the principles that persisted from the Cold War, and ultimately motivated the Outer Space Treaty[1] and its subsequent counterparts,[2] are no longer sufficient. Truth be told, they were never expected to be so. The United Nations Committee on the Peaceful Uses of Outer Space (“COPUOS”) never contemplated commercial uses when it adopted—and many nations subsequently ratified—its longstanding space treaties. While private actors have interacted with this environment for decades, the commercial space industry has only recently reached a point of maturity where entities can productively utilize orbital environments, cultivate an entirely new source of natural resources in lunar and cislunar space and further explore the translunar realm. Commercial space is having its moment, and it represents a monumental paradigm shift for space law and policy.

Considering the radical evolution of actors and activities in space, do the instruments and institutions that oversee it need to evolve as well? Traditional forms of public international lawmaking—multilateral treatymaking and institution building followed by each participant’s cooperative consent—may not meet the needs of private actors who bear little affiliation to the country they select to license their operations. Similarly, domestic regulations and policies from a government-mission minded era appear ill suited for the novel complexities of the commercial launch and communications capabilities that are rapidly eclipsing those of national governments. The diverse set of actors and activities in outer space also introduce a novel set of contexts and conflicts that impact private law. In effect, commercial space activity is spurring change that no one track can resolve independently, necessitating pluralist reform that extends the bounds of both public and private law.

A second-order problem that emerges is how to manage an ecosystem in which collective commercial interests diverge from national interests. As many nations become dependent on commercial space services and infrastructure, the balance of power is shifting toward a new calculus. Decisions by private actors now impose externalities that national actors experience immediately and directly, and vice versa, making both sides of the public-private dichotomy increasingly intertwined. Thus, if the law is intended to evolve into more efficient, wealth-maximizing rules, we must also ask who reaps the benefits of these efficiencies, and do they lead to sound policy?

These questions are vexing but timely and provide ample room for further scholarly development exploring ways to better manage the use of outer space. On February 3, 2023, the Journal of Law & Innovation hosted its symposium, “The Emerging Commercial Space Age: Legal and Policy Implications” at the University of Pennsylvania Carey School of Law.[3] The Symposium brought together leading international law scholars, economists, and telecommunications and antitrust policymakers to assess the twenty-first century space domain and its implications for legal and policy frameworks. Panelists and moderators emphasized the progress of commercial enterprise in outer space, how these increasingly complex and multifaceted interests would influence international space law and the paradigm shifts that must emerge in economic regulation and public policy to foster innovation and sustainable competition. The Articles in this volume touch each of these considerations and are an outgrowth of the presentations and moderated discussions at the Symposium.

[1] Treaty on Principles Governing the Activities of States in the Exploration and Use of Outer Space, Including the Moon and Other Celestial Bodies, Jan. 27, 1967, 18 U.S.T 2410, 610 U.N.T.S. 205 (entered into force Oct. 10, 1967) [hereinafter Outer Space Treaty].

[2] Agreement on the Rescue of Astronauts, the Return of Astronauts and the Return of Objects Launched into Outer Space, Apr. 22, 1968, 19 U.S.T. 7570, 672 U.N.T.S. 119; Convention on the International Liability for Damage Caused by Space Objects, Mar. 29, 1972, 24 U.S.T. 2389, 961 U.N.T.S. 187 [hereinafter Liability Convention]; Convention on Registration of Objects Launched into Outer Space, Nov. 12, 1974, 1023 U.N.T.S. 15; Agreement Governing the Activities of States on the Moon and Other Celestial Bodies, Dec. 5, 1979, 1363 U.N.T.S. 3 [hereinafter Moon Agreement].

[3] The symposium program and webcasts of the presentations and discussions are available at https://www.law.upenn.edu/institutes/ctic/jli/events.php.

Abstract

Accelerating demand for wireless broadband is accentuating the need to optimize use of limited spectrum resources that are susceptible to congestion. Recent technological innovations enable exclusive-use, licensed spectrum and open-access, unlicensed spectrum to serve as complementary goods. We present a game-theoretic model in which wireless broadband service providers engage in simultaneous pricing and service decisions for a heterogeneous consumer population. We demonstrate that for some unlicensed allocations, service providers may maximize profit by offloading some consumer traffic onto the unlicensed band. Consequently, adding unlicensed capacity can increase congestion in wireless spectrum bands in ways that harm social and consumer welfare. These effects are reminiscent of Braess’s Paradox, in which adding capacity counterintuitively leads to greater congestion. Notably, these effects emerge through supply-side differentiation strategies, rather than demand-side responses. We then utilize our framework to analyze recent high-profile decisions by the FCC and introduce a framework for identifying the appropriate balance between licensed and unlicensed allocations.

Abstract

China is making an active push to enlarge its role in the development of Internet-related technical standards. The prevailing narrative surrounding this trend suggests that Beijing is aiming to uproot the liberal, democratic values embedded in the Internet’s technical foundation and governance arrangements in favor of authoritarian-friendly alternatives. For many, these fears were fully realized when Chinese tech giant Huawei came to the UN-affiliated International Telecommunications Union (ITU) and proposed the development of a future core Internet protocol called “New IP”. This proposal allegedly sought to redesign the architecture of the Internet in a way that would both enhance and export the Chinese government’s capacity for digital repression. Informed by the understanding of Chinese standards influence as a geopolitical and ideological threat, many are now calling for a more aggressive response to countering Chinese engagement in Internet standards bodies.

Yet, the conventional narrative seems to be missing something. Specifically, it overlooks the fact that the sophisticated Internet control apparatus China has developed over the years can already censor and surveil quite effectively at present and that shifting responsibility for core protocol development to the state-driven ITU would not necessarily enhance its ability to do so. A more comprehensive understanding of this trend is needed.

Using New IP as the primary case study, this article examines China’s standard-setting push, its potential motivations, and its implications for the future of the global Internet. We conclude that it is far from clear that New IP was indeed intended as a trojan horse for digital authoritarianism. Observing that technical evolution of the Internet—particularly the type endorsed in Huawei’s proposal—plays a prominent role in China’s long-term industrial policy strategy, we find it equally plausible that New IP was motivated by economic considerations, something that has largely been absent from the debate over China’s standards ambitions. We thus caution against the presumption that Chinese-developed standards are intended to advance the cause of digital repression as well as against politically driven opposition to growing Chinese participation at Internet standard-setting bodies. This insight is crucial, as the way American policymakers and Internet stakeholders respond to this trend will undoubtedly impact both the future of the global Internet and U.S. technological leadership in this domain.

Abstract

Whether network slicing complies with the net neutrality rules currently in force in Europe and previously applicable in the U.S. presents a key issue in the deployment of 5G. In many ways, both regimes frame the issues in a similar manner, with the exceptions for reasonable traffic management and specialized services likely to play the most important roles. Both regimes also focus on similar considerations, including the requirement that measures be based on technical rather than business considerations and the distinction between measures aimed at improving the performance of the entire network or specific applications, although both distinctions are problematic in some respects. Both regimes also emphasize application agnosticism and end-user choice, with European law finding the former implicit in the latter. At the same time, European and U.S. law reflect some key differences: the regimes cover different types of entities, frame the issues in terms of nondiscrimination versus throttling and paid prioritization, take different positions on whether measures must be limited to temporary or exceptional circumstances, and place different weight on the impact of the rules on investment and on the relevance industry standards. The relatively undeveloped state of both legal regimes means that the ultimate answer must await enforcement decisions and actions by NRAs, and any subsequent judicial challenges to these decisions.

Abstract

Although most studies of major communications reform legislation focus on the merits of their substantive provisions, analyzing the political dynamics behind the legislation can yield important insights. An examination of the tradeoffs that led the major industry segments to support the Telecommunications Act of 1996 (the “1996 Act”) provides a useful illustration of a political bargain. Analyzing the current context identifies seven components that could form the basis for the next communications statute: (1) universal service; (2) pole attachments; (3) privacy; (4) intermediary immunity; (5) net neutrality; (6) spectrum policy; and (7) antitrust reform. Assessing where industry interests overlap and diverge and the ways that the political environment can hinder passing reform legislation provides insights into how these components might combine to support the enactment of the next Telecommunications Act of 1996.

Abstract

The growing economic importance of technical standards has heightened the need for a better understanding of why they succeed or fail. While existing literature has scrutinized the role of public governance, particularly in the realms of regulation, antitrust, and intellectual property, to date legal scholars have largely overlooked the role of private organizational and contractual lawyering in determining the path of technical standardization.

In this Article, we explore this dimension through a case study of the effects of private organizational governance and contracting practices on the fortunes of a nascent Internet security standard. The standard, known as Resource Public Key Infrastructure (“RPKI”), is designed to increase the trustworthiness of information about Internet routing. Through analysis of private organizational and contractual documents, semi-structured interviews with participants in the Internet operations industry, and attendance and participation in key industry conferences, we gained an embedded perspective on the role that private lawyering played in shaping would-be adopters’ perceptions and decisions regarding the technical standard.

According to our interviewees, contract and organizational bureaucracy mattered greatly. Notably, we found that the terms of contractual agreements prevented some potential adopters from experimenting with the technology and deterred others from proposing that their organizations adopt the technology. This was due to the perceived costs of involving organizational lawyers in technology-adoption decisions. In addition, contract terms deterred actors from increasing the functional value of the standard via complementary innovation and the development of complementary information services. Remarkably, even the basic mechanisms for presenting and assenting to contract terms chilled prospects for adoption. Regarding organization, we found that stark differences of governance and mission between key North American and European nonprofits contributed to different patterns of adoption. Taken together, these findings reveal the continuing importance of old-school transaction-cost engineering even in the most technical realms of Internet operation and standardization.