
By Jeffrey T. Donner, Esq.
September 23, 2026
Section 230 of the Communications Decency Act is often described in shorthand as the statute that protects websites from liability for content posted by their users. That description is directionally correct, but incomplete. The harder question—and the one that has occupied courts for decades—is where the protection ends.
A website may provide a forum in which users publish defamatory statements and still be protected from liability. It may review those statements before publication, select some and reject others, edit them for length or clarity, and even maintain a business model built around controversial user submissions. None of those facts necessarily makes the website responsible for the defamatory content.
But Section 230 does have a boundary. A website that becomes responsible, “in whole or in part,” for the creation or development of the unlawful information becomes an “information content provider” as to that information. 47 U.S.C. § 230(f)(3). The difficult cases therefore ask not merely whether the website hosted, encouraged, or benefited from controversial material, but whether the website itself materially contributed to what made the material unlawful. The distinction remains one of the most important—and sometimes counterintuitive—features of Internet defamation law.
That question has particular significance to me because nearly two decades ago I represented a plaintiff in litigation involving one of the more provocative websites of the early social-media era: DontDateHimGirl.com.
The Statutory Framework
Section 230(c)(1) provides that “[n]o provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider.” Congress simultaneously declared a policy favoring both the development of Internet services and private efforts to screen objectionable content. The statute therefore protects not only passive conduits but, in many circumstances, websites exercising the traditional editorial functions of a publisher. The current statutory language remains materially the same on this central point.
That rule was in significant part a congressional response to an early pair of cases that created a peculiar incentive for online services not to moderate their content.
In Cubby, Inc. v. CompuServe Inc., 776 F. Supp. 135 (S.D.N.Y. 1991), an online service that exercised little editorial control over third-party content received treatment resembling that of a distributor rather than a publisher. Several years later, however, a New York trial court in Stratton Oakmont, Inc. v. Prodigy Services Co., 1995 WL 323710 (N.Y. Sup. Ct. May 24, 1995), treated Prodigy as a publisher in part because it had undertaken to moderate its bulletin boards.
That produced an obvious perverse incentive. A service that ignored objectionable material might face less exposure than one that attempted to police it. Section 230 was enacted against that background. The Florida Supreme Court later discussed precisely this history in Doe v. America Online, Inc., 783 So. 2d 1010 (Fla. 2001), adopting a broad understanding of Section 230’s protection against state-law liability based on third-party content.
The result is important for any modern website operator. Moderation, standing alone, does not generally destroy Section 230 protection. A website does not become the author of a user’s defamation merely because it decides whether the statement should appear.
From Publisher Liability to “Development” of Unlawful Content
The key statutory term is “information content provider.” Section 230 defines that phrase as a person or entity “responsible, in whole or in part, for the creation or development” of information provided through the Internet. 47 U.S.C. § 230(f)(3).
The word “development” has therefore done considerable doctrinal work.
Courts have rejected two extremes. On one side, Section 230 cannot mean that a website loses protection whenever it exercises editorial judgment, because that would recreate the problem Congress was attempting to solve. On the other side, a website cannot manufacture unlawful content itself and escape liability merely because a user supplied some of the underlying information.
The prevailing judicial formulation is the “material contribution” test. A website becomes responsible for development when it materially contributes to the aspect of the content that makes it unlawful.
The Ninth Circuit’s en banc decision in Fair Housing Council of San Fernando Valley v. Roommates.com, LLC, 521 F.3d 1157 (9th Cir. 2008), is the landmark example. Roommates.com required users to disclose information concerning protected characteristics and required them to state preferences used in matching prospective roommates. The court concluded that Section 230 did not protect the website from claims arising from information that the site’s own structure required users to provide in an allegedly discriminatory manner. By contrast, the website remained protected with respect to a free-form “Additional Comments” section in which users independently chose what to write.
That distinction is critical. Merely giving users a place to speak is one thing. Designing the interaction in a way that materially contributes to the particular illegality is another.
Soliciting Complaints Is Usually Not Enough
The Fourth Circuit confronted a related argument in Nemet Chevrolet, Ltd. v. ConsumerAffairs.com, Inc., 591 F.3d 250 (4th Cir. 2009). A Chevrolet dealership alleged that ConsumerAffairs.com solicited complaints, organized them into categories, and participated sufficiently in the resulting content to become an information content provider.
The court rejected the claim. Even assuming the website solicited complaints and organized user submissions, the allegations did not plausibly establish that it contributed to the allegedly fraudulent or defamatory nature of the statements themselves. The distinction was not whether the website encouraged criticism. It was whether the website was responsible for what allegedly made the particular statements unlawful.
That principle substantially narrows the argument that a website loses Section 230 protection simply because its entire business model invites accusations.
A restaurant-review website plainly solicits criticism of restaurants. A consumer-complaint site invites dissatisfied customers to accuse businesses of misconduct. A dating-warning site may invite users to describe alleged infidelity or mistreatment. The fact that criticism is foreseeable—or even the principal reason the site exists—does not automatically convert the website operator into the creator of every defamatory statement a user chooses to publish.
The Dirty.com Decision and the Limits of “Encouragement”
The Sixth Circuit made the point even more forcefully in Jones v. Dirty World Entertainment Recordings LLC, 755 F.3d 398 (6th Cir. 2014).
TheDirty.com was hardly a neutral electronic bulletin board. It operated as a gossip website. Users submitted salacious accusations about identifiable people. The site’s operator selected submissions for publication and sometimes added his own commentary. A jury found the operator liable for defamatory user submissions, and the district court concluded that the website had forfeited Section 230 protection by encouraging actionable posts.
The Sixth Circuit reversed.
Adopting the material-contribution standard, the court held that encouraging offensive or defamatory submissions in a general sense was not enough. The operator had selected the posts for publication, but he had not materially contributed to the defamatory content within the particular submissions at issue. His own commentary was attributable to him, of course, but it did not transform him into the author of the user’s accusations merely because he endorsed, mocked, or reacted to them.
That decision illustrates just how substantial Section 230 protection can be. A website can be created specifically to attract scandalous allegations, knowingly publish hostile accusations about real people, and engage editorially with those accusations without necessarily becoming liable for the third party’s defamatory statements.
That does not mean the operator has absolute immunity. It means the plaintiff must identify the website’s own contribution to the unlawful content rather than relying merely on the site’s atmosphere, motives, or editorial choices.
Where the Line Can Be Crossed
Other cases demonstrate the opposite result.
In FTC v. Accusearch Inc., 570 F.3d 1187 (10th Cir. 2009), the defendant operated a website through which customers could obtain confidential telephone records. The operator paid researchers to obtain the information and sold the resulting data to customers. The Tenth Circuit concluded that the operator had participated in developing the unlawfully obtained information rather than merely serving as a passive intermediary.
Similarly, Roommates.com demonstrates that a website can lose protection where its own mandatory questions, categories, or architecture materially produce the allegedly illegal result.
The conceptual boundary therefore is not best described as “moderated versus unmoderated,” or even “neutral versus biased.” The better distinction is between facilitating publication and materially participating in the illegality.
A website normally remains protected when a user independently decides to say:
“John Smith cheated on me.”
The analysis becomes more difficult if the website requires the user to answer:
“Which sexually transmitted disease did John Smith knowingly give you?”
The first situation provides a forum. The second may embed an allegedly defamatory factual premise into the site’s own question and direct the user toward supplying a particular accusation. Whether that would ultimately defeat Section 230 would depend on the facts and governing circuit law, but it illustrates why the design of the platform can matter.
Likewise, if the operator rewrites a user’s submission to add a defamatory allegation, fabricates facts, combines separate submissions to create a new defamatory implication, or personally supplies the actionable statement, Section 230 does not immunize the operator’s own content.
DontDateHimGirl.com and an Early Section 230 Fight
Those principles bring the issue back to Hollis v. Cunningham, No. 1:07-cv-23112-CMA (S.D. Fla.).
DontDateHimGirl.com was an early Internet site that permitted women to post photographs and information about men they believed other women should avoid dating. Contemporary accounts described profiles accusing men of infidelity and other misconduct. The site was operated by Tasha Cunningham, formerly known as Tasha Joseph. (CBS News)
Pittsburgh attorney Todd Hollis became the subject of postings on the site. He initially sued in Pennsylvania, but the claims against the Florida defendants were dismissed for lack of personal jurisdiction. In November 2007, Hollis filed a federal action in the Southern District of Florida asserting defamation, false-light invasion of privacy, and intentional infliction of emotional distress. I later served as Florida counsel for Hollis in that litigation. The case was assigned to U.S. District Judge Cecilia M. Altonaga.
The case squarely implicated Section 230.
The defendants contended that they could not be held liable for content supplied by third-party users. In April 2008, they moved for partial summary judgment, asserting among other arguments that portions of the challenged material were protected by Section 230. Hollis opposed the motion. The court never ultimately decided the Section 230 question. The case ended on June 20, 2008, when the parties stipulated to dismissal with prejudice following settlement.
The practical objective of the litigation was achieved without producing an appellate precedent. But the legal issue remains unusually instructive.
Was DontDateHimGirl.com Merely a Forum?
The strongest argument for Section 230 protection was straightforward: the allegedly defamatory accusations had been written by users, not by the website operator.
The opposing theory was more subtle. DontDateHimGirl.com was not a general-purpose message board where defamation happened incidentally. The central purpose of the site was to invite women to publish negative information about identifiable men for the benefit of prospective romantic partners.
Indeed, during the earlier controversy surrounding Hollis, Cunningham publicly described the site as a forum intended to allow women to warn each other, while Hollis argued publicly that the site was actively encouraging women to submit malicious and defamatory information. Contemporary reporting reflects that dispute. (CBS News)
At an intuitive level, there is an obvious difference between Facebook unknowingly hosting a defamatory post among billions of unrelated communications and a website specifically constructed for people to publish accusations against identified individuals.
But Section 230 doctrine does not turn primarily on that intuition.
Under the later decisions in Nemet and Jones, the mere fact that a website solicits negative information—and even builds its audience around such information—generally would not establish that the operator developed the defamatory content. The plaintiff would need evidence connecting the operator to what made the particular statement false and defamatory.
That is a much more demanding showing.
What Would Matter Under Modern Law
If the same case were litigated today, discovery into the site’s architecture and editorial practices would be especially important.
Relevant questions could include whether the operator merely supplied an open text box or instead asked leading factual questions; whether users were required to select accusations from predetermined categories; whether the operator edited submissions substantively rather than cosmetically; whether site personnel suggested allegations to users; whether they combined accusations from different sources; whether they added headlines that themselves communicated defamatory factual assertions; whether they knew material had not actually been submitted for publication by the purported author; or whether they otherwise contributed factual content of their own.
An operator’s motive to attract sensational material would not by itself resolve the Section 230 issue. Neither would proof that inflammatory content increased traffic or advertising revenue. Those facts might explain the site’s business incentives, but the statutory inquiry remains whether the defendant was responsible for the creation or development of the unlawful information.
That is where Section 230 sometimes produces results that conflict with ordinary notions of editorial responsibility.
Under traditional defamation law, a newspaper that knowingly republishes someone else’s defamatory accusation ordinarily does not escape responsibility merely because another person wrote the original words. Section 230 deliberately alters that common-law rule for covered Internet intermediaries.
Florida’s Broad Application of Section 230
Florida courts have long recognized that broad protection.
In Doe v. America Online, Inc., the Florida Supreme Court held that Section 230 barred a negligence theory seeking to impose liability upon AOL for third-party content despite allegations that AOL had received complaints about the underlying activity. The court expressly relied on Zeran v. America Online, Inc., 129 F.3d 327 (4th Cir. 1997), one of the foundational federal cases interpreting Section 230.
The Eleventh Circuit likewise has recognized the distinction between third-party authorship and content created by the defendant. In Whitney Information Network, Inc. v. Xcentric Ventures, LLC, 199 F. App’x 738 (11th Cir. 2006), an unpublished decision involving operators of a consumer-complaint website, the court observed that allegations that defendants themselves authored statements could make them information content providers rather than merely publishers of another person’s content.
The principle is therefore familiar in Florida litigation: immunity depends heavily on who was responsible for the challenged content.
Section 230 Is Powerful, but It Is Not a License to Defame
Section 230 is sometimes discussed as though it abolished defamation liability on the Internet. It did not.
The person who actually creates a defamatory statement remains potentially liable under ordinary law. A website remains responsible for its own statements. A website also can become responsible for third-party information to the extent that it participates sufficiently in creating or developing what makes the information unlawful. And Congress has enacted specific statutory exceptions to Section 230, including provisions addressing certain sex-trafficking claims. 47 U.S.C. § 230(e).
What Section 230 largely prevents is the transformation of an Internet intermediary into the legal publisher of somebody else’s words merely because it hosted, selected, moderated, refused to delete, or otherwise exercised ordinary editorial control over them.
That remains a remarkable departure from traditional publication law.
The Enduring Problem
The DontDateHimGirl.com litigation arose during an earlier Internet era, before Facebook, X, Reddit, TikTok, and algorithmically curated social media acquired their present significance. Yet the central legal question has hardly disappeared.
Indeed, modern platforms make the question more important.
The more sophisticated a platform becomes, the harder it can be to distinguish between merely transmitting user speech and actively shaping it. Recommendation engines choose what users see. Prompts determine what users are asked to say. Algorithms amplify certain subjects. Interfaces channel users into predefined categories. Artificial-intelligence systems can rewrite, summarize, combine, or generate material based on user submissions.
At some point, “hosting” shades into “development.”
Section 230 does not supply a simple answer to where that point lies. What it does provide is a framework. The central inquiry is not whether a website moderates content, whether it profits from controversy, or whether its operators should have known that users would make defamatory accusations. The inquiry is whether the website itself became responsible, at least in part, for the creation or development of what made the challenged information unlawful.
Nearly twenty years after Hollis v. Cunningham, that remains the line lawyers and courts must find. And as Internet platforms become increasingly active participants in organizing, prompting, editing, and generating online speech, the distinction between publishing another person’s words and helping create them may become harder—not easier—to draw.
