I.     Introduction & Summary

As society continues to adapt to the digital world, the Federal Communications Commission (FCC) stands at the forefront to facilitate the development and deployment of necessary broadband infrastructure that makes this transition possible. The FCC has taken this role to heart, consistently removing outdated regulations that impede technological advancement.[1] Over the last four years, the FCC has also worked to undo past regulatory overreach, ensuring that it remains within its statutory authority granted by Congress.[2] Indeed, the efforts by the FCC have helped pave the way for a massive increase in online traffic and the growth of a wide variety of services that consumers use to check news, find information and stay connected with friends.[3]

But with all advancement comes growing pains. In recent years, many have begun to worry about the role that social media plays in the national conversation. Disinformation campaigns using social media seek to destabilize the American public.[4] Terrorist organizations use Internet platforms to recruit new members.[5] And many worry that bias among social media platforms directly targets specific ideologies, though both sides disagree on which viewpoints. [6]

Unsurprisingly, these concerns have led to calls for action.[7] Because Section 230 immunizes platforms from legal suit, this provision sits in the sights of lawmakers and pundits alike.[8] However, calls for reform often ignore the important role that protections play in allowing companies to remove these categories of content while also allowing the free expression of ideas.

The Petition for Rulemaking makes this same mistake. In attempting to solve a problem that may or may not even exist, the National Telecommunications and Information Administration (NTIA) petition would derail the entire intermediary liability regime that has made American technology companies the envy of the world, while drastically increasing the FCC’s authority of the Internet.[9] Further, as the legislative history and the statutory text makes clear, the FCC has no legal authority to interpret the statute as the NTIA asks.

These comments argue that the FCC should not proceed with a notice of proposed rulemaking. First, the comments explain the history of intermediary liability protections in the United States and the role that Section 230 plays in resolving what is known as the “moderator’s dilemma.” Second, the specific interpretations that the comments seek from the FCC would cause real harms to speech online. Finally, even if the proposal in the petition wouldn’t cause the exact harms Section 230 seeks to prevent, the FCC lacks the legal authority from Congress to interpret Section 230’s provisions.

II.     Section 230 Remains a Vital Protection for Speech Online

Strong, intermediary liability protections allowed the Internet to grow into the transformative tool for communications it has become.[10] Unfortunately, many calls for reform, including this petition, fail to adequately understand how and why these protections evolved into the current regime we have today, as well as the effect that removing or limiting them would have. As a result, the reforms promise to do much more harm that than potential issues they seek to address. As the FCC contemplates the future of Section 230, it must do so with a complete understanding of why the statute remains vital to protect speech online.

A.    History of Intermediary Liability in the United States

To understand why strong, intermediary liability protections remain vital for online platforms, it is first important to understand the issues that arose prior to Section 230 and how courts applied case law to the Internet.

Intermediary liability refers to the liability for content an intermediary hosts or otherwise provides to the consumer, though that party does not generate the content itself. Often times, this comes in the form of a bookseller or a newspaper stand, in which the intermediary provides a variety of products to the consumer, but does not publish the content they sell.

Indeed, common law developed under this exact scenario. The most important case dealt with a bookseller who sold a book called Sweeter than Life, which arguably violated a California provision that prohibited the possession of obscene material.[11] When challenged, the court needed to determine whether the bookseller could be held liable for the obscene content of the book under the statute.[12] As the court reasoned, the local ordinance’s lack of a scienter requirement threatened to dramatically limit free speech, as booksellers who had no knowledge that a particular book contained obscene material would still violate the ordinance.[13] For instance, without a scienter element, a bookseller would not be able to sell a wide array of books without reading every single book that passed through their stores. This meant that the available texts, regardless of whether they actually contained obscene material, would be unduly restricted by the ordinance. Therefore, the court decided to examine whether the bookseller had actual knowledge of the obscene content before liability spread to the seller.[14]

In the context of physical goods like books and newspapers, this analysis makes perfect sense. However, as courts began to grapple with the application of this standard to online platforms, issues arose. In Cubby v. Compuserve, the Southern District of New York had to determine whether Compuserve, and more importantly its online general information service, could be liable for specific content that an online newsletter published.[15] As New York generally required knowledge of the infringing content before a plaintiff could hold an intermediary liable for that content, the court ultimately examined whether Compuserve exercised any control over the content.[16] In other words, it examined whether Compuserve moderated any of the content that users posted to its online general information service.[17] Because Compuserve took no such action, the court determined they didn’t have the requisite knowledge of defamatory content posted by users and thus the company couldn’t be held liable.[18]

This analysis fit squarely in line with Smith v. California and the scienter examination. However, so too did the analysis in another case: Stratton Oakmont v. Prodigy.[19] Unlike Compuserve, Prodigy did moderate the content that users posted to a bulletin board on its service.[20] One user posted defamatory content about Stratton Oakmont, but because the plaintiffs couldn’t identify the individual user, they targeted Prodigy for hosting the content. Because the service did moderate user content, the court established the “Prodigy Exception.”[21] Under this analysis: “[I]f an online publisher exercised control over third-party content, then the provider was a publisher, not a distributor that was entitled to the protections of Cubby and Smith.”[22]

This left platforms in a dilemma: either refuse to moderate to ensure the platform would not have any knowledge of defamatory content, or moderate and remove any content that could potentially lead to litigation. Unfortunately, this “moderators dilemma” would leave users worse off regardless of what the platform decided to do.

B.     Section 230 Resolves the “Moderators Dilemma”

Section 230 was the Congressional response to this dilemma. Social media amplifies voices and connects people across the globe, and while the benefits are clear, like any tool it can be used for nefarious ends. Users, and more importantly advertisers, will not use a service if it contains harmful content that the users do not wish to see or advertisers do not want associated with their product. However, if services potentially open themselves up to liability because they take steps to remove the harmful content, they may refuse to moderate at all. Alternatively, the service may over-remove user speech or eliminate user-generated content entirely, taking away the significant benefits that social media provides to communities across the globe.[23]

Congress resolved this problem by developing a law which encouraged platforms to moderate the content on their services without fear of liability.[24] The law, which was passed as Section 230 of the Communications Decency Act, was included with an omnibus telecommunications reform package, the Telecommunications Act of 1996.[25] The statute has two primary components: It prohibits treating an interactive computer service as the publisher or speaker of user-generated content regardless of the editorial discretion the service takes, and encourages platforms to develop their own moderation standards.[26]

The first component, 47 USC § 230(c)(1) has been the primary focus of the courts as it provides interactive computer services immunity from litigation for the content their users.[27] The second component, 47 USC 230(c)(2), receives less attention but works in tandem to 230(c)(1).[28] Instead of blanket immunity, Congress designed this provision to encourage platforms to set their own content moderation standards.[29] These two provisions incentivize platforms to moderate content without fear that such moderation will lead to liability for their standards or for the content that remains on the service. If users or advertisers dislike the moderation practices of a given service, then they can choose to use a different service. This means that each individual service will craft policies that best meet the demands of their users.

Exceptions to Section 230 do exist, most notably in the context of intellectual property and the enforcement of federal crimes .[30] However, these protections remain a vital means for platforms to target harmful content while keeping speech they feel comfortable hosting.

Calls for reform often attempt to frame the statute as a giveaway to interactive computer services that do not exist for other forms of media.[31] But this framing misses the purpose of Section 230. It was not a sweetheart deal for the largest players or subsidies to grow their Internet businesses—instead, it allows all websites, regardless of size or design, to develop services by which users can share information. And if we are to reform the law, it is critical that we do not recreate the same dilemma the law sought to resolve.

III.     The Proposed Interpretations Contravene the Intended Goals Congress Sought to Achieve by Passing Section 230

Even assuming the FCC has the authority to interpret Section 230—which it does not, as Section IV will explain—the proposed changes would ignore the purpose of the statute and ultimately recreate the same issues Congress passed Section 230 to prevent.

A. The Interaction Between Subparagraphs 230(c)(1) and 230(c)(2)

First, the NTIA asks the FCC to declare that:

47 U.S.C. 230(c)(1) applies to an interactive computer service for claims arising from failure to remove information provided by another information content provider. Section 230(c)(1) has no application to any interactive computer service’s decision, agreement, or action to restrict access to or availability of material provided by another information content provider or to bar any information content provider from using an interactive computer service. Any applicable immunity for matters described in the immediately preceding sentence shall be provided solely by 47 U.S.C. § 230(c)(2).[32]

Essentially, the NTIA wants to shift more authority to 230(c)(2)(A) because it includes the terms “good faith” and “otherwise objectionable,” which could theoretically provide an interpretive light for the FCC. However, this is not the case, and it contravenes the purpose of the statute to limit the applicability of 230(c)(1).

As previously explained, Congress passed Section 230 to ensure that the threat of litigation would not disincentive moderation efforts. The two provisions of Section 230 work together to make clear to the courts that a company’s decision to moderate content cannot be used as evidence that they are liable under traditional, intermediary liability case law. As Jeff Kosseff has explained:

(c)(1) and (c)(2) mean that companies will not be considered as the speaker or publishers of third-party content, and they will not lose that protection only because they delete objectionable posts or otherwise exercise good-faith efforts to moderate user content.[33]

Unfortunately, the NTIA’s proposal would contravene this purpose. There is a reason that many defendants rely on the protections of 230(c)(1) when available: it is a blanket immunity that can assert at motion to dismiss phase. And indeed, as the 4th Circuit explained:

Specifically, § 230 precludes courts from entertaining claims that would place a computer service provider in a publisher’s role. Thus, lawsuits seeking to hold a service provider liable for its exercise of a publisher’s traditional editorial functions — such as deciding whether to publish, withdraw, postpone or alter content — are barred.[34]

This absolute protection makes it cheaper and quicker to defend under 230(c)(1) than it is to use 230(c)(2)(A).[35] By trying to limit the applicability of 230(c)(1), interactive computer services will face additional costs defending lawsuits and may alter their moderation practices accordingly.

B.     The Meaning of Section 230(c)(2)

Next, the NTIA petition asks the FCC to interpret the terms “good faith” and “otherwise objectionable.”[36] An incorrect reading of the statute could theoretically give the FCC some leeway to define the terms.

However, when examining the history of the statute, it is clear that the drafters wanted to give platforms leeway to define the types of content that they want to allow on their service. This is because different platforms are designed with different users in mind, and what is objectionable to some may be reasonable to others. With strong competition online, Section 230’s design allows these services to tailor their standards to provide a unique experience that attracts users.[37]

Clearly, courts have understood the importance of allowing services to design their own moderation standards. While good faith is generally an objective standard, courts have held “otherwise objectionable” to be subjective.[38] This allows platforms to freely define what they find to be objectionable, and so long as they conclude that content is objectionable, the ‘good faith’ piece does not meaningfully restrict their behavior.

If the FCC attempts to limit the ability for platforms to design their own moderation policies, it runs the risk of preventing companies from targeting harmful content that does not fit within one of the three categories defined in the petition. Worse, content that may fit one of the categories may not be prohibited by platforms if they worry that their subjective decisions do not meet the FCC’s definition of otherwise objectionable, or that excessive litigation will derive from these decisions.

This will ultimately leave consumers worse off, as platforms may be driven to allow content that they otherwise would not. Section 230 strikes a delicate balance; the NTIA’s proposals to redefine “otherwise objectionable” and “good faith” will upset this balance and limit free speech online.

C.     Section 230(c)(1) and 230(f)(3)

Ultimately, Section 230 only covers user-generated content. This means that Section 230 has no bearing on any content generated by the interactive computer service, such as the fact check on the president’s tweet which ultimately led to the executive order.[39] However, the NTIA seeks to expand what it means for an interactive computer service to generate content. They suggest the following interpretation:

For purposes of 47 U.S.C. § 230(f)(3), “responsible, in whole or in part, for the creation or development of information” includes substantively contributing to, modifying, altering, presenting or prioritizing with a reasonably discernible viewpoint, commenting upon, or editorializing about content provided by another information content provider.[40]

The NTIA relies  on Fair Housing Council of San Fernando Valley v. Roommates.Com to support its assertions.[41] In that case, Roommates.com specifically asked the users’ demographic information and allowed others to find roommates (in part) based on these demographics.[42] However, the Fair Housing Act  (FHA) expressly prohibits discrimination on the basis of some of these demographics. As a result, the 9th Circuit concluded that Section 230 did not protect Roommates.com, which specifically had a category selection for a roommates race.[43] Four years later, the court determined that in-fact the actions of the website didn’t violate the FHA, but regardless the interpretation on what it means for an interactive computer service to generate the content has stood: “If you don’t encourage illegal content, or design your website to require users to input illegal content, you will be immune.”[44]

Appropriately, this interpretation has been narrowly tailored. If interactive computer services are generating content, then Section 230 should not apply. But only if the interactive computer service is in-fact generating that content. Roommates.com, for example, required users to provide their demographic information and fill out questionnaires.[45] This isn’t the case for most services, and merely moderating or removing content is distinct from generating that content. And indeed, modifying and removing content is specifically the type of action that Congress sought to encourage when it passed Section 230, and courts have consistently determined that traditional publisher actions will not lead to liability.[46]

The interpretation supported by the NTIA would effectively destroy Section 230. Almost all current action by interactive computer services would theoretically fall under this interpretation. For example, if a text chat app automatically filters out specific words and censors them, the app developers would no longer be able to avail themselves of the protections of 230(c)(1) because they are ultimately modifying user content. And if the FCC’s interpretation covers removal, it would directly contradict the statute. If not, then platforms will likely remove the content entirely rather than blocking specific words or phrases and risk litigation for every piece of content that remains on the platform. Worse, the platform may remove the chat feature altogether.

These outcomes pale in comparison to the impact of interpreting the generation of content to include prioritizing content with a “reasonably discernable viewpoint.”[47] Clearly, this targets the supposed anti-conservative bias that many accuse technology platforms of holding. Yet there is no evidence such bias exists, and some studies suggest the opposite.[48]

The NTIA admittedly does not appear to seek to prohibit prioritization outright, and instead aims to limit the interpretation to cases when there is a discernable viewpoint.[49] However, what constitutes prioritizing a discernable viewpoint will likely end up a question that requires litigation, and one with no clear definition.[50] With the fear of litigation consistently in the background, platforms may remove prioritization features altogether, making it more difficult for users to find content and suppressing usage across the board.

To some, that could be the ideal outcome, but this viewpoint is short-sighted and ignores the impact on the user. Most notably, platforms will likely remove more conservative speech—the very speech that the Executive Order seeks to protect—because many conservative posts could lead to litigation if the platform does not moderate in a politically neutral way. Platforms may choose to remove entirely legitimate content that would otherwise remain online to avoid the appearance that moderation efforts had a “discernable viewpoint” or only remove content from one perspective. While this jawboning of private speech directly contradicts the first amendment—ignoring the fact that all speech online will suffer—it may also lead to the suppression of the very speech the president claims he wants to protect.[51]

D.    Treated as Publisher or Speaker

Finally, the petition calls on the FCC to define: “treated as the publisher or speaker of any information provided by another information content provider.”[52] In the NTIA’s view, an interactive computer service should lose the protection of 230(c)(1) if it affirmatively solicits content or if the service affirmatively vouches for, editorializes, recommends or promotes the content on the basis of message.[53]

This proposal would drastically limit the user experience. First, many platforms are designed for specific purposes or points of discussion. For example, Reddit allows users to create their own communities about a given topic.[54] Arguably, these communities actively solicit content regarding those subjects. But they need to do so, as this solicitation creates the type of community that the users are looking for. These individual communities are distinct because they only allow a specific subject content, and in-turn the communities grow and develop their own norms. However, the NTIA’s interpretation of Section 230 threatens to remove protections for platforms that solicit specific types of content, meaning that these interactive computer services may remove these communities if they will lead to increased litigation risks.

Likewise, the petition’s prohibition on recommending or promoting content based on that content’s substance will cause significant harms.[55] Platforms recommend and promote content similar to content that a user interacts with on the service. For example, if an individual watches a video on Roman wars with Vercingetorix and Boudicca, YouTube will try to find other videos that relate to similar topic such as the Punic Wars or Antony’s Civil War, or vice versa.[56] However, if recommending videos could potentially lead to future litigation, platforms may decide to remove the recommended feature altogether. This means that users will struggle to find new content that would interest them, and it would make it more difficult for new content creators to attract an audience.

The NTIA tries to alleviate these concerns by creating carveouts, yet these carveouts fail to understand the importance of Section 230 as outlined above.[57] For example, the NTIA suggests that a “good faith application of [the service’s] terms of service” would not lead to treatment as a publisher or speaker.[58] But whether a recommendation or a solicitation is a good faith application of the terms of service is again a litigable question. Section 230 works because it takes the risk of litigation out of the equation when determining whether to moderate content.[59] If the platform faces constant litigation because users do not think that the platform was acting in good faith or if it complied with the terms of service, then the service may simply choose to forego the recommendations or the solicitation of specific content entirely to avoid any potential litigation. This is especially true of new or nascent services who do not have the resources to compete with incumbent services.

The petition consistently tries to solve problems with apparent anti-conservative bias with changes to Section 230. However, the proposed solutions in the petition do not address these concerns because the petition ignores the Prodigy Exception and why Section 230 became law. Instead of a meaningful solution, the petition seeks changes to Section 230 as a means to punish companies for acting in a way the president dislikes. While this is a blatant attempt to contravene the First Amendment, it is also bad policy that will stifle free speech and competition in the online space.

IV.     The FCC lacks the statutory authority to interpret Section 230

This market-based approach to addressing harmful content online has worked well, making American technology companies the envy of the world. Now, the president wants the FCC to step in and take additional authority to regulate a new sector of the economy. However, Congress designed the statute to guide courts in litigation against interactive computer services, and the FCC has no role in interpreting the clear language and purpose of the statute.

A.  Section 230 was designed to guide courts and grants no authority to the FCC to interpret its provisions

As explained in Section II, the drafters of Section 230 designed the statute to resolve a problem in the courts. The language of the statute clearly tracks this legislative goal, providing no role for any administrative agency to interpret its statute. And unfortunately for the FCC, “the question a court faces when confronted with an agency’s interpretation of a statute it administers is always, simply, whether the agency has stayed within the bounds of its statutory authority.”[60]

First, in the general findings and policy sections, Congress makes clear that the purpose of the statute was simply to guide courts. The congressional finding in Section 230(a)(4), for example, states: “[T]he Internet and other interactive computer services have flourished, to the benefit of all Americans, with a minimum of government regulation.”[61] And in the policy section, Congress states the purpose of the law is: “[T]o preserve the vibrant and competitive free market that presently exists for the Internet and other interactive computer services, unfettered by Federal or State regulation.”[62] Taken together, these two provisions make clear that Section 230’s approach to resolving the moderator’s dilemma envisioned no role for the administrative state to step in and regulate content online.

More importantly, the language of the statute provides no role for administrative interpretation. Instead, it clearly lays out a standard for reviewing courts to adhere to. In 230(c)(1), the statute clearly lays out a judicial standard: interactive computer services will not be treated as the publisher or speaker of user-generated content.[63] This unambiguous statement of the law directly targets legal actions against companies like Compuserve and Prodigy, and explains that courts cannot hold platforms liable for what the users post. Similarly, the drafters designed 230(c)(2)(A) to encourage platforms to establish their own internal standards for content and moderation to oppose a more heavy-handed approach to regulating the Internet.[64] When viewed in the context of working with 230(c)(1) to design a regime in which the market addresses harmful speech online, it becomes clear that the drafters envisioned no role for the FCC in its implementation.

This becomes even more evident when looking through the legislative history. In the record, numerous statements from the drafter highlight the importance of keeping the FCC out of the role of content moderator:

Some have suggested, Mr. Chairman, that we take the Federal Communications Commission and turn it into the Federal Computer Commission, that we hire even more bureaucrats and more regulators who will attempt, either civilly or criminally, to punish people by catching them in the act of putting something into cyberspace.[65]

If we regulate the Internet at the FCC, that will freeze or at least slow down technology. It will threaten the future of the Internet. That is why it is so important that we not have a Federal computer commission do that.[66]

Congress did not want the FCC to step in and regulate the Internet. It made this abundantly clear by designing the statute to self-execute and lay-out the findings and purpose at the forefront. If a change to the law is needed, Congress must be the entity to make that change.

B.     The FCC lacks the authority to interpret Section 230 under its general rulemaking authority

Clearly, Congress granted no authority to the FCC in the plain text of Section 230. However, the petition argues that the FCC can use its general rulemaking authority to interpret Section 230.[67] The FCC has general rulemaking authority to interpret the laws under its jurisdiction.[68] Because Section 230 and the Communications Decency Act passed as a part of the Telecommunications Act of 1996, Section 230 does fall within the FCC operative statutes.[69] However, for the FCC to use its general rulemaking authority, there must be some delegation authority from Congress.[70] NTIA tries to get around this by arguing that because the statute is technically in their jurisdiction—and because they have general rulemaking authority—the FCC can resolve ambiguities in the statue.[71] Unfortunately for the NTIA, even if the statute didn’t unambiguously prevent the FCC from interpreting its provisions, the provisions lack the necessary ambiguity the petition seeks to manipulate.

At the outset, the NTIA tries to argue that there is some ambiguity regarding how Section 230(c)(1) and Section 230(c)(2)(A) interact.[72] Unfortunately, as courts have consistently made clear, the protections provided by 230(c)(1) cover a vast array of editorial functions such as the removal or modification of content: there is no ambiguity in its application.[73] When viewing the statute from the context of resolving the Prodigy Exception and countering the overhanded approach of the Communications Decency Act, the two provisions work together to establish a regime in which platforms will not be liable for user-generated content and moderating content will not give rise to liability. The petition tries to bridge the two sections because 230(c)(1) is so explicit in its language that any attempts to find ambiguity will ultimately fail, which is explained below.

Next, the petition tries to redefine “good faith” and “otherwise objectionable” in 230(c)(2)(A).[74] Even if “good faith” was ambiguous, “otherwise objectionable” is an entirely subjective standard, and courts have treated it as such.[75] This means that as long as the platform objects to the content, it is otherwise objectionable.[76] In other words, there is no ambiguity for the FCC to resolve. Services can define their own standards for otherwise objectionable, as Congress envisioned when it took this hands-off approach to content moderation.

Third, the petition seeks to find ambiguity in what it means to generate content, which again, fails.[77] The plain meaning of terms like “creation” and “development” do not extend to actions such as the removal of posts, modification of harmful content or prioritizing information in any way. In fact, despite the petitions claims otherwise, courts have consistently determined that 230(c)(1) covers withdrawal or altering content.[78] This is because they are traditional editorial functions, which Section 230 was designed to not only allow, but actively encourage. Even Fair Housing Council of San Fernando Valley v. Roommates.Com, the case which the petition relies on, was narrowly tailored to a company that provided questionnaires with potentially discriminatory categories for finding a roommate, not things like an additional comment feature.[79] The case law clearly supports this unambiguous definition of content generation.

Finally, in trying to find ambiguity in the phrase “treated as the publisher or speaker” of user-generated content, the petition again ignores the plain meaning of the statute.[80] On its face, the statute plainly exempts platforms from treatment as the publisher or speaker of what a user posts, regardless of whether the platform recommends or prioritizes that content. The statute was designed to allow services to moderate and control user-generated content without being liable for it. That was the entire point of its passage. By ignoring this to arbitrarily argue that prioritization or recommendation of such content removes Section 230 protections for the service, the petition seeks to add additional language to the statute that is not there, which clearly exceeds the FCC’s authority.

And to be clear, the FCC has already made the determination that Section 230 provides no authority for it to interpret its provisions. In the Restoring Internet Freedom Order, the FCC consistently cited Section 230 as proof that Congress did not intend for them to regulate the Internet, and instead takes a hands-off approach.[81] As the FCC explained: “Section 230(b) is hortatory, directing the [FCC] to adhere to the policies specified in that provision when otherwise exercising our authority.”[82]

In other words, the efforts by the petition to find ambiguity in the statute fail because the statute clearly envisioned no role for an administrative agency to interpret its provisions. The petition tries to find the ambiguity as a means for getting around this fact, but the language of the statue is clear and cannot be rewritten just because the president is mad that his content violates terms of service.

The FCC has consistently worked to allow markets to efficiently allocate resources and determine the best policies. This was the goal of the drafters of Section 230. The FCC should not forego its recent work simply because the president dislikes how private companies are responding to the content he produces. Not only would this be a flagrant attack on the First Amendment and a blatant regulatory power grab which future administrations can exploit, it would also be bad policy.

Respectfully submitted,


Jeff Westling

Technology and Innovation Resident Fellow

1212 New York Avenue NW, Suite 900

Washington, D.C. 20005

Sept. 2, 2020

[1] See, e.g., Declaratory Ruling and Third Report and Order, In the Matter of Accelerating Wireless Broadband Deployment by Removing Barriers to Infrastructure Investment et al., WT Docket Nos. 17-79 and 17-84, (Sept. 27, 2018). https://bit.ly/2TP3hFQ; Order on Reconsideration and Second Notice of Proposed Rulemaking, In the Matter of 2014 Quadrennial Review—Review of the Commission’s Broadcast Ownership Rules and Other Rules Adopted Pursuant to Section 202 of the Telecommunications Act of 1996 et al., MB Docket No. 14-50 et al. (Nov. 16, 2017). https://bit.ly/36Gj3G7.

[2] See, e.g., Declaratory Ruling, Report and Order, and Order, In the Matter of Restoring Internet Freedom, WC Docket No. 17-108 (Dec. 14, 2017). https://bit.ly/3aG532M [RIF Order].

[3] “Internet/Broadband Fact Sheet,” Pew Research Center, June 19, 2019. https://pewrsr.ch/3iTpImI.

[4] “Report of the Select Committee on Intelligence of the United States Senate on Russian Active Measures Campaigns and Interference In the 2016 U.S. Election,” Volume 2: Russia’s Use of Social Media with Additional Views, 116th Congress (Oct. 09, 2019). https://bit.ly/2Q96TPW.

[5] Antonia Ward, “ISIS’s Use of Social Media Still Poses a Threat to Stability in the Middle East and Africa,” The RAND Blog, Dec. 11, 2018. https://bit.ly/2Eka30A.

[6] Shannon Bond, “Trump Accuses Social Media of Anti-Conservative Bias After Twitter Marks His Tweets,” NPR, May 27, 2020. https://n.pr/2Yh1I4X.

[7] President Donald J. Trump, “Executive Order on Preventing Online Censorship,” May 28, 2020. https://bit.ly/328WXvs.

[8] Casey Newton, “Everything You Need to Know About Section 230,” The Verge, May 28, 2020. https://bit.ly/3aFe1Nu.

[9] Petition For Rulemaking of the National Telecommunications and Information Administration, RM-11862 (July 27, 2020) [Petition]. https://www.ntia.gov/files/ntia/publications/ntia_petition_for_rulemaking_7.27.20.pdf.

[10] See, e.g., Jeff Kosseff, The Twenty-Six Words That Created the Internet (Cornell University Press, 2019).

[11] Smith v. California, 361 US 147 (1959).

[12] Id.

[13] Id. at 150-53.

[14] Id.

[15] Cubby, Inc. v. Compuserve Inc., 776 F. Supp. 135 (S.D.N.Y. 1991). https://bit.ly/319kfSE.

[16] Id. at 139-40.

[17] Id.

[18] Id.

[19] Stratton Oakmont, Inc. v. Prodigy Services Co., Index No. 94-031063 (N.Y. Sup. Ct., Nassau County May 24, 1995).

[20] Kosseff, supra note 10, at 49.

[21] Id. at 51-52.

[22] Id. at 50-51.

[23] “Submitted Statement for the Record of Jeffrey Westling, Kristen Nyman and Shoshana Weissmann Before the House of Representatives Communications and Technology and Consumer Protection and Commerce Subcommittees of the Energy and Commerce Committee,” Hearing on Fostering a Healthier Internet to Protect Consumers Section 230 of the Communications Decency Act, 116th Congress (Oct. 16, 2019). https://bit.ly/327PVqP.

[24] Kosseff, supra note 10 at 64.

[25] Telecommunications Act of 1996, Pub. L. No. 104-104 (1996).

[26] Kosseff, supra note 10 at 64-65.

[27] Id.

[28] Id.

[29] Id. at 65.

[30] See 47 U.S.C. 230(e)(2); 47 U.S.C. 230 (e)(1).

[31] Sen. Josh Hawley, “President Trump Praises Sen. Hawley’s ‘Ending Support For Internet Censorship Act,’” Press Release, July 11, 2019. https://bit.ly/2QdOkds.

[32] Petition at 31.

[33] Kosseff, supra note 10, at 66.

[34] Zeran v. America Online, Inc., 129 F.3d 327, 330 (1997).

[35] Eric Goldman, “Trump ‘Preventing Online Censorship’ Executive Order Is Pro-Censorship Political Theater,” Technology & Marketing Law Blog, May 29, 2020. https://bit.ly/3aJWfbP.

[36] Petition at 37-38.

[37] Alec Stapp, “10 Myths About Big Tech and Antitrust,” Progressive Policy Institute, July 28, 2020. https://bit.ly/2Q7Y60K.

[38] Goldman, supra note 37.

[39] Comments of Ashken Kazaryen, “James Madison Institute Town Hall on Encryption,” at 42:33 (Aug. 13, 2020). https://www.youtube.com/watch?v=UPOcty4tZOM&feature=youtu.be&t=2552.

[40] Petition at 38-39.

[41] Fair Housing Council of San Fernando Valley v. Roommates.com, LLC, 521 F.3d 1157 (9th Cir. 2008). https://scholar.google.com/scholar_case?case=7987071093240934335&q=roommates+&hl=en&as_sdt=20006.

[42] Id. at 1169.

[43] Id.

[44] Eric Goldman, “Roommates.com Isn’t Dealing in Illegal Content, Even Though the Ninth Circuit Denied Section 230 Immunity Because It Was,” Technology Law and Marketing Blog, Feb. 6, 2012 [quoting Fair Housing Council of San Fernando Valley v. Roommate.com, LLC,] 2012 WL 310849 (9th Cir. Feb. 2, 2012). https://bit.ly/34hs9LQ.

[45] Fair Housing Council of San Fernando Valley v. Roommates.com, supra note 41, at 1169.

[46] Zeran v. America Online, Inc., 129 F.3d 327, 330 (1997).

[47] Petition at 42.

[48] Mike Masnick, “New Study Finds No Evidence of Anti-Conservative Bias In Facebook Moderation,” Techdirt, June 2, 2020. https://bit.ly/32nqaDa.

[49] Petition at 42.

[50] See “Section 230: Cost Report,” Engine, 2019. https://bit.ly/2Ew1IqY.

[51] Jeffrey Westling, “Hawley’s Attack on Section 230 Hurts Conservatives,” R Street Institute (June 24, 2019). https://www.rstreet.org/2019/06/24/hawleys-attack-on-section-230-hurts-conservatives/.

[52] Petition at p. 45; 47 U.S.C. 230(c)(1).

[53] Petition at p. 45.

[54] Jake Widman, “What Is Reddit?”, Digital Trends, July 1, 2020. https://bit.ly/2Yue6Pe.

[55] Petition at 46.

[56] See generally “Product Features: Recommended videos,” YouTube Aug. 25, 2020. https://www.youtube.com/howyoutubeworks/product-features/recommendations.

[57] Petition at 47.

[58] Id.

[59] Eric Goldman, “Why Section 230 Is Better than the First Amendment,” 95 Notre Dame L. Rev. 33 (2019).

[60] City of Arlington v. FCC, 133 S. Ct. 1863. 1868 (2013).

[61] 47 U.S.C. § 230(a)(4).

[62] 47 U.S.C. § 230(b)(2).

[63] 47 U.S.C. § 230(c)(1).

[64] Kosseff, supra note 10, at 64.

[65] 141 Cong. Rec. H8469 (Aug. 4, 1995). https://www.govinfo.gov/content/pkg/CREC-1995-08-04/pdf/CREC-1995-08-04-pt1-PgH8460.pdf.

[66] Id. at H8471.

[67] Petition at 15.

[68] See 47 U.S.C. § 154(i); 47 U.S.C. § 201(b); 47 U.S.C. § 303(r).

[69] Harold Feld, “Could the FCC Regulate Social Media Under Section 230? No,” Public Knowledge, Aug. 14, 2019. https://bit.ly/34xmzoA.

[70] Id. See also Motion Picture Ass’n of America v. Fed. Comm. Comms’n., 309 F.3d 796 (D.C. Cir. 2002). https://bit.ly/32rc9UL.

[71] Petition at 15.

[72] Id. at 30-31.

[73] See also Zeran v. AOL, 129 F.3d 327, 330 (4th Cir. 1997).

[74] Id. at 37-40.

[75] Goldman, supra note 37.

[76] Id.

[77] Petition at 42.

[78] See also Zeran v. AOL, 129 F.3d 327, 330 (4th Cir. 1997).

[79] Fair Housing Council of San Fernando Valley v. Roommates.com, 521 F.3d 1157 (2008). https://scholar.google.com/scholar_case?case=7987071093240934335&q=roommates+&hl=en&as_sdt=20006.

[80] Petition at 46-47.

[81] See generally RIF Order

[82] Id. at 284.