If web publishers were held responsible for information and comments posted by other people, the internet as we know it would cease to exist. We would never have known companies like Facebook or Twitter, as social media consists almost entirely of information provided by others. So too with all message boards, Craigslist, Wikipedia and so on.

Congress recognized early on that the growth of a vibrant and robust internet required protection for web publishers from lawsuits arising from content posted or provided by someone else. It did so with Section 230 of the Communications Decency Act. Section 230(c)(1) provides that no web publisher (which the statute calls an “interactive computer service”) shall be deemed the publisher or speaker of information provided by another person (i.e., another “information content provider”). Thus, no web publisher may be held liable as a “publisher or speaker” of offending information provided by another “information content provider.” This immunity is so airtight that plaintiffs, for the most part, no longer even bother to try to hold website publishers liable for content posted by someone else.

But a web publisher can sacrifice this immunity by getting too involved in the production of the offending content. Section 230 immunity only applies so long as the web publisher itself is not also an “information content provider” with respect to the offending content. The statute defines an “information content provider” to be “any person or entity that is responsible, in whole or in part, for the creation or development” of the offending content. 47 U.S.C. Section 230(f)(2).

The CDA does not define “responsible,” “create” or “develop.” While the term “create” is straightforward enough, the tougher and more ambiguous issues arise regarding whether the publisher is partially responsible for “developing” content. Web publishers should be mindful of at least three situations where courts have held Section 230 moot because of the web publisher’s own responsibility for developing the offending content.

First, a web publisher should not expect immunity where it has taken affirmative steps to solicit the offending content that was provided by another person. In FTC v. Accusearch, 570 F.3d 1189 (2009), Accusearch provided a service that allowed users to obtain detailed, confidential phone records for the mobile or landline numbers of other people. Accusearch would take an order, assign it to an outside researcher/investigator, pay that researcher directly and then provide the data to the customer via its website.

Accusearch argued it was not responsible either for creating or developing the offending content. The phone numbers and records were data “created” by the phone companies, defendant argued, and only outside researchers worked to obtain that data. But the court held that Accusearch read those terms much too narrowly. The court explained the term “create” had to mean something more than simply “to bring into existence” or else the term “develop” would be redundant. The court delved into the etymological origins of the term “develop” and found the crux of the term involved the act of drawing something out, making it “visible,” “active” or “usable.” Under this standard, Accusearch had developed the offending information because it had taken purely private data and effectively weaponized it by making it available to any paying customer on the Internet. Clearly, this was a transformation of that private data. It further held that Accusearch was “responsible” at least in part for the information, a test which could be met “only if it in some way specifically encourages development of what is offensive about the content.” The term thus had a moral connotation in the sense of being responsible for the offensive nature of the information. Because Accusearch knew that it was paying the researchers to obtain purely private data that it knew could only be obtained through fraudulent or illegal means, the company was far from a morally blameless, neutral conduit.

Second, web publishers may be deemed to “develop” defamatory comments on their site when they participate in the authorship of the offending content. In Huon v. Denton, 841 F.3d 733 (7th Cir. 2016), the 7th U.S. Circuit Court of Appeals held that a plaintiff successfully pleaded around Section 230 immunity by alleging plausibly that a defamatory comment in response to an article about his rape trial was not written by a third-party reader, but actually by Gawker staffers. Gawker allegedly had become the “information content provider” with respect to a post that stated the defendant had committed the crime despite being acquitted for it.

Nothing on its face suggested that a Gawker staffer had written this comment. But the plaintiff backed up his theory with several pages of factually detailed allegations to support why Gawker would plausibly encourage — if not author — a defamatory comment such as this. Those allegations “hinted at why Gawker employees might have anonymously authored comments,” to wit: to generate more buzz about the article, increase readership and increase paid advertisements. Plaintiff also cited several passages from a news article about how Gawker was trying to “monetize” its comments to attract more advertising dollars.

But without this type of strong factual foundation, a plaintiff should not expect to plead around Section 230 by alleging defendant’s authorship of ostensible third party content. Unfounded assertions in this regard will fail the Iqbal and Twombly pleading standards and will face dismissal. In Kimzey v. Yelp, 836 F.3d 1263 (9th Cir. 2016), the court affirmed an order dismissing a complaint as barred by Section 230, where the plaintiff, a locksmith, alleged Yelp had actually authored the one-star review referenced in his complaint. The court held that a plaintiff cannot offer threadbare allegations of the website’s authorship of defamatory content; allowing a plaintiff to do so would contravene Congress’ intent in including Section 230.

Third, a web publisher may be held to be responsible for “developing” offending content that the site requires as a condition of using the site. In Fair Housing Council of San Fernando Valley v. Roommates.com, LLC, 521 F.3d 1157 (9th Cir. 2008), the 9th U.S. Circuit Court of Appeals held that Roommates.com had “developed” the content that potentially violated the Fair Housing Act by requiring users to disclose roommate preferences that involved proscribed selection criteria such as sex, sexual orientation, and whether they intended to bring children into the home. The court rejected Roommates.com’s Section 230 immunity argument that the preference information consisted solely of content provided by third parties. The site had set up a drop-down menu from which users had to choose from among the unlawful criteria. The website required users to submit this information as a precondition for using the services. The court analogized this feature to a real estate broker who told a potential client, “Tell me whether you are Jewish or I will not show you this listing.” Due to the company’s role in requiring users to make those choices, the court had no problem holding that Roommates.com, at least in part, “developed” the offending content.

Notably, the court held that the website was immune from liability stemming from comments posted in a space provided to users to explain in their own words what attributes the user was seeking in a roommate. In this scenario, the website played no role in the users’ choice of content and thus the website was not a “content provider” of that information.

Section 230 remains an extremely powerful shield against liability for web publishers, but only for those that heed the lessons from cases where Section 230 did not apply: Remain neutral as to third-party information. Do not preselect what content may be posted. And if you’re paying for information to appear on your site, make sure the source of that data is legitimate.

For further information about this or related matters, please contact Christian P. Foote at 650-342-9600 or cfoote@carr-mcclellan.com.