More Section 230 Cases Than I Can Handle!
My cup runneth over with Section 230 cases! This long blog post catches up on a few from the past couple months. Warning: there are some stinkers in this batch.
Google, Inc. v. Hood, 2015 WL 1546160 (S.D. Miss. March 27, 2015)
As you know, for some time Mississippi Attorney General Jim Hood has been haranguing Google about third party content on Google’s network, which culminated with a 79 page subpoena from Hood’s office to Google apparently predicated on legal theories that bear no resemblance to the actual law. Google challenged this subpoena in court, and the resulting opinion contributes to the canon of First Amendment and Section 230 cases protecting search engines. Ironically, although it’s not exactly how he envisioned it, Jim Hood is helping make legal principles that benefit his constituents and is garnering big headlines along the way.
The Section 230 discussion is murky. The court cites Google’s self-policing efforts and the 2013 letter to Congress sent by the state attorneys general complaining about how Section 230 restricts their authority (that was in the context of their anti-online prostitution advertising efforts). But the court doesn’t clearly explain why these facts support the TRO.
The court’s First Amendment discussion is much clearer:
Google’s publishing of lawful content and editorial judgment as to its search results is constitutionally protected. See Jian Zhang v. Baidu.com Inc., No. 11 CIV. 3388 JMF, 2014 WL 1282730 (S.D.N.Y. Mar. 28, 2014)(“there is a strong argument to be made that the First Amendment fully immunizes search-engine results from most, if not all, kinds of civil liability and government regulation”)(citations omitted). The Attorney General’s interference with Google’s judgment, particularly in the form of threats of legal action and an unduly burdensome subpoena, then, would likely produce a chilling effect on Google’s protected speech, thereby violating Google’s First Amendment rights.
To me, this reasoning applies equally to a wide swath of state attorneys general enforcement actions against Internet companies.
The court continues:
it is well-settled that the Attorney General may not retaliate against Google for exercising its right to freedom of speech by prosecuting, threatening prosecution, and conducting bad-faith investigations against Google
Ouch! “BAD FAITH”!
Google also gets great rulings on the Fourth Amendment…
Attorney General Hood’s subpoena must comport with the requirements of the Fourth Amendment and not wage an unduly burdensome fishing expedition into Google’s operations.
…and FDCA preemption…
Google points out that the subpoena demands information concerned with Google’s dealings with Canadian online pharmacies in violation of federal law. This court, then, finds it necessary to temporarily enjoin Attorney General Hood’s enforcement of the subpoena until the court resolves the extent, if any, to which the demands contained therein are preempted [by the Food, Drug and Cosmetics Act].
…and copyright preemption:
The subpoena contains various requests for information regarding copyright infringement. Many of these requests are found in a section titled “Stolen Intellectual Property”. It is well-established that state attorneys lack the authority to enforce the Copyright Act; such enforcement power lies with the federal government.
The court also cites Section 512’s applicability, but like the Section 230 discussion, the court’s thought process is incomplete.
I hope Mississippi voters will keep this ruling in mind during the next election cycle. I’m struggling to see how Hood’s stated objectives advance his constituents’ interests at all, and I doubt even more how those objectives became a high enough priority to trump the many other important functions that state AGs should be performing. (Well, actually, from the Project Goliath emails, we all know that Hood has disquieting ties to the MPAA, but that also seems like helpful information for Mississippi voters to know.)
Faegin v. LivingSocial Inc., 2015 WL 1198654 (S.D. Cal. March 16, 2015):
This is a troubling ruling. The principal players are “A.T. Your Service Cleaning” and its rival “At Your Service Housekeeping.” The pseudo-trademark problems should be obvious: both parties adopted virtually identical but highly descriptive, if not generic, business names. There is a good chance these so-called trademarks won’t survive this litigation.
The A.T. enterprise ran LivingSocial promotions for a few months; then its rival ran LivingSocial promotions. The A.T. enterprise claimed that its rival failed to do a good job for its customers and didn’t provide its contact information prominently enough, and the resulting consumer frustration and confusion led to unwarranted bad consumer reviews and hurt the A.T. enterprise’s business.
The A.T. enterprise also sued LivingSocial for running the bad ad promotion for the rival. LivingSocial sought to dismiss the claims on Section 230 and other grounds. LivingSocial points out that the A.T. enterprise’s real beef is with the rival’s business name, and the business names are third party content to LivingSocial. However, the complaint repeatedly alleged that LivingSocial “partnered” with the rival, so the court concludes that it can’t dismiss on Section 230 grounds:
The FAC alleges that Defendant LivingSocial advertises and sells the allegedly misleading vouchers. From this allegation, the Court is able to draw the “reasonable inference” that Defendant LivingSocial was “‘responsible, in whole or in part’ for creating or developing” the content made available on LivingSocial’s website.
It’s really hard to figure out what to make of this conclusion. One argument is that LivingSocial is basically an ad network, in which case it should easily qualify for Section 230 for third party ads. But we’ve also seen some troubling Section 230 rulings for ad networks, including the Swift v. Zynga and Chang v. Wozo rulings (both involving AdKnowledge) as well as the LeanSpa ruling discussed below. So one possibility is that courts are choking on Section 230 for ad networks, at least at early stages of the litigation.
Another way of reading the ruling is that LivingSocial was the retailer in this equation, so it faces first-party liability for the goods and services it sells even if those are ultimately fulfilled with third party vendors. That interpretation would comport with cases like the uncited FTC v. AccuSearch case, but it doesn’t resolve how this case differs from the many Section 230 wins by eBay and Amazon for their marketplaces.
The opinion goes on to talk about the 1114(2)(B) trademark safe harbor for innocent printers and publishers. We rarely see interpretations of that statute, but sadly, the 1114 discussion is as cryptic as the Section 230 discussion. The court says:
The allegations of the FAC do not demonstrate that Defendant LivingSocial is entitled to protection under the Lanham Act’s safe harbor provision as an innocent infringer” or “innocent violator.” 15 U.S.C. § 1114(2)(B). The FAC alleges that Defendant LivingSocial was aware of A.T. Your Service Cleaning and Janitorial’s mark from their prior partnership and then advertised with At Your Service Housekeeping in the same market.
I continue to believe that Section 1114 safe harbor is an important component of online intermediary liability scheme, but this ruling doesn’t support that hypothesis.
FTC v. LeanSpa, 2015 WL 1004240 (D. Conn. March 5, 2015).
This case has been troubling me ever since I first blogged it 2 years ago. I wrote last time that the ruling sucked, and that’s how I feel about the latest opinion too.
The case relates to the numerous fake news websites created to promote the health benefits of acai berries. Some of those sites participated in an affiliate network run by LeadClick for the benefit of the advertiser, LeanSpa. Let’s assume for now that both LeanSpa and the creators of fake news sites violated FTC law. What about the liability of the advertising network in the middle, LeadClick?
On the surface, this sounds like an easy Section 230 win. Any problems with the content of either LeanSpa or the fake news website creators would necessarily hold LeadClick responsible for third party content. Yet, the court sides with the FTC in this case. How?
The court starts with the prima facie case of a Section 5 violation. LeadClick is held to the following legal standard:
Courts have held individual defendants liable for a corporation’s conduct where they “(1) participated in the acts or had authority to control the corporate defendant and (2) knew of the acts or practices.”
This is a contributory copyright-esque standard: knowledge + control/participation. LeadClick met this standard with respect to the fake news site operators.
Knowledge: “it is undisputed that LeadClick employees knew that fake news sites were being used to promote LeanSpa products on the eAdvertising Network.” Notice the narrow view of scienter in the test: it doesn’t require knowledge of any legal violation. And of course every ad network will generally know what types of content its network publishers are providing.
Control: the court says (cites omitted):
no reasonable jury could deny that LeadClick both participated in, and had the authority to control, the affiliate marketers conduct in so far as it related to the fake news sites. Specifically, LeadClick’s affiliate managers were tasked with scouting for new affiliates. Affiliate managers “routinely gathered information about affiliates,” and they would solicit affiliates to join eAdvertising. Thus, LeadClick solicited and hired affiliate marketers using fake news sites to advertise LeanSpa’s products. Notably, “[a]ffiliate marketers had to apply to join the eAdvertising Network, and LeadClick would decide which to accept.” Indeed, “the standard contract that governed LeadClick’s relationship with its affiliates” states:
All websites, newsletters, companies, or individuals need official approval from eAdvertising before they can become a member of the Publisher Program. Only websites and newsletter that have been reviewed and approved are permitted to use the programs. eAdvertising reserves the
right to withhold or refuse approval on any website, newsletter, company, or individual for any reason, whatsoever.
Although LeadClick asserts that, “in practice, affiliate marketers were not required to submit their ‘websites’ for approval unless [it] specifically requested to see the page,” it does not deny that it had the authority to review pages. Indeed, after the FTC began suing affiliate marketers in April 2011 for using fake news sites, LeadClick started to screen fake news pages by removing their ability to advertise certain products without approval from the merchant. This evidence establishes that, as a matter of law, LeadClick had the authority to control the affiliate marketers’ use of fake news pages…
LeadClick had the authority to not hire affiliates using fake news sites, to instruct them not to use such sites after hiring them, and to remove them if they continued to do so. Just as LeanSpa would be liable for approving requests to advertise with fake news sites, LeadClick, as LeanSpa’s agent, is liable for its own decision to effectuate that decision.
In other words, LeadClick ran an advertising network. Ad networks have to police their network; it’s not only required by their advertisers, but increasingly they are under legal compulsion to do so (see, e.g., copyright, CA’s law on advertising to minors). Thus, the court’s analysis should equally apply to any other ad network.
The court continues that LeadClick participated in the deception based on the following evidence (cites omitted):
LeadClick purchased advertising space on genuine news sites and sold it to affiliates advertising with fake news sites. In doing so, it provided a way for consumers to browse directly from a genuine news site to a fake one. Indeed, it is clear that LeadClick knew it was creating such a bridge between genuine and fake news sites because, on some occasions, LeadClick would identify fake news pages “as destination pages for the banner ads when negotiating with media sellers,” and sometimes it “identified the destination webpage for the banner ads by emailing the media seller a compressed version of the affiliate’s” fake news page.
LeadClick also discussed product pairings that appeared on the fake news sites. Specifically, the fake news pages contained a purported reporter’s test of a two-step product combination. LeadClick participated in the fake news sites’ deception by, for example, “implementing a rule that publishers pair . . . [LeanSpa] products with other [LeanSpa] products” and going “through all the publishers that [were] running [a LeanSpa] offer and then tell[ing] them that they needed to run the new step 2.”
I don’t fully understand these points. To me, it sounds like what ad networks do.
Having established LeadClick’s prima facie violation, the court then turns to the Section 230 defense (cites omitted):
No reasonable jury could deny that LeadClick was an “information content provider.” LeadClick solicited and hired the affiliate marketers to advertise LeanSpa’s products, knowing that affiliates used fake news pages. LeadClick continued paying Davidson and other affiliates running fake news pages for their referrals to LeanSpa’s website. LeadClick communicated with LeanSpa and with affiliates running fake news pages regarding which products should be advertised as the “Step 1” and “Step 2” products purportedly under independent investigation. LeadClick also screened advertisements according to merchants’ preferences.
LeadClick’s media buying also materially contributed to the unlawful nature of the fake news sites by providing affiliates running fake news sites with a way to direct consumers from genuine news sites to fake news sites. The alleged deception in this case is the representation that independent testing was being conducted by genuine news reporters; LeadClick’s media buying contributed to that deception by providing consumers with yet another reason to think that the news site was genuine.
I don’t even know what to say about this. The “analysis” apparently conflates the prima facie elements with the immunity defense. Plus, the court compounds the Section 230 problem by holding the ad network accountable for the content on the legit news sites in addition to the fake news sites.
Unquestionably, fake news sites are bad news. No judge is going to be sympathetic to them. Ad networks shouldn’t be doing business with fake news sites. But this ruling goes way too far in bending both the prima facie case and Section 230 to tag the intermediary. In light of the other bad ad network rulings, we seem to be moving towards a SOPA-esque world where every ad network is fully responsible for the content on network publishers’ sites, meaning they will dump any publisher than poses any legal risk to them.
A final stinger: the court holds CoreMetrics, LeadClick’s parent, accountable as a relief defendant for another $4M of damages.
Opperman v. Path, 2015 WL 1306494 (N.D. Cal. March 23, 2015)
Jones, the one new case cited by Apple, does not change the Court’s conclusion that Apple is an “information content provider.” The Jones court merely adopted the Ninth Circuit’s reasoning in Roommates.com, 521 F.3d 1157 at 1168, that, if the alleged content provider is not a creator of the challenged content, it must have done more than merely “encourage” the creation of the challenged conduct; the alleged provider must have required another to create that content. ECF No. 501 at 26-27 (quoting Jones, 755 F.3d at 414). The Court previously considered this portion of Roommates, and determined that Plaintiffs’ allegations were sufficient at this stage to survive a CDA challenge.
Here, Plaintiffs allege that Apple effectively was a creator, at least in part, of the App Defendants’ Apps that misappropriated Plaintiffs’ address book data. ECF No. 478, ¶ 46 (“all iDevice Apps were built, in part, by Apple.”). Plaintiffs have also provided factual detail regarding how Apple controlled the development of Apps provided by its App Store, and contributed to the creation of the offensive content, for example, by using its iOS Human Interface Guidelines to instruct App developers to access address book data without prior permission. This is all the information-content-provider exemption requires: “The term ‘information content provider’ means any person or entity that is responsible, in whole or in part, for the creation or development” of the offending content. Neither Jones nor the new allegations in the SCAC compel the Court to reverse its prior conclusion.
Apple, as an alleged “information content provider” under the CDA, is not entitled, as a matter of law and at this stage of the proceedings, to blanket CDA immunity for the conduct challenged here.
Kabbaj v. Google Inc., 2015 WL 534864 (3d Cir. Feb. 10, 2015)
Kabbaj filed a complaint in the District Court against Google, Inc., Amazon, Inc., Yahoo, Inc., and ten “John Doe” defendants, charging defamation, tortious interference with contract, and negligent and intentional infliction of emotional distress based on various online postings. The District Court, in a comprehensive opinion, properly held that Kabbaj’s claims against Google, Amazon, and Yahoo are barred by the Communications Decency Act, 47 U.S.C. § 230(c)(1), (e)(3). See Green v. America Online (AOL), 318 F.3d 465, 470–71 (3d Cir.2003) (Act provides immunity to interactive computer service providers “as a publisher or speaker of information originating from another information content provider”).
Westlake Legal Group v. Yelp, 14-1872 (4th Circuit March 18, 2015)
the facts alleged in the complaint and attached exhibits indicate, at most, that Yelp has an automated system that filters reviews. Such activities constitute traditional editorial functions that do not render Yelp an information content provider.
Tobinick v. Novella, 2015 WL 1191267 (S.D. Fla. March 16, 2015):
Upon a careful review of the briefing, the exhibits, and the record as a whole, the Court concludes that whether or not the Society is responsible for the content of the articles, neither article constitutes commercial speech, at least as regards the Society.
That disputed fact is material with respect to the Society’s argument that they are entitled to CDA immunity on Plaintiffs’ defamation claims. For CDA immunity to apply, the Society must be a “provider or user of an interactive computer service.” Whitney Info. Network, Inc. v. Xcentric Ventures, LLC, No. 204–CV–47–FTM–34SPC, 2008 WL 450095, at *7 (M.D.Fla. Feb. 15, 2008). Because the relationship between the Society and the SBM Blog remains in dispute, the Court cannot ascertain whether the Society is such a provider with respect to the SBM Blog.
Followup ruling from April denying the request for a preliminary injunction.
Nestle Purina PetCare Co. v. Blue Buffalo Co. Ltd., 2015 WL 1782661 (E.D. Mo. April 20, 2015):
PRCG/Haggerty briefly argues that it is immune from liability under the Communication Decency Act (“CDA”), 47 U.S.C. § 230 et seq., because Congress “made the legislative judgment to effectively immunize providers of interactive computer services from liability in tort with respect to material disseminated by them but created by others.” See Blumenthal v. Drudge, 992 F.Supp. 44, 49 (D.D.C.1998). PRCG/Haggerty argues that under the CDA, it cannot be liable for the content Purina or anyone else provided for the website. Blue Buffalo responds that PRCG/Haggerty’s argument fails because (1) CDA immunity is an affirmative defense that a plaintiff is not required to plead around, and (2) PRCG/Haggerty is an “information content provider” and accountable under the CDA for the content it created for the Honesty website.
Under the CDA, “the term ‘information content provider’ means any person or entity that is responsible, in whole or in part, for the creation or development of information provided through the Internet or any other interactive computer service.” 47 U.S.C. § 230(f)(3). The CDA does not protect internet “content providers” from responsibility for the content they transmit through the web. See, e.g., Hy Cite Corp. v. Badbusinessbureau.com, 418 F.Supp.2d 1142, 1146, 1148–49 (D.Ariz.2005) (allegations that defendant produced original content and editorials and created titles to the defamatory reports posted by users of the website were sufficient to survive motion to dismiss). Blue Buffalo is correct that it is not required to plead around affirmative defenses, and PRCG/Haggerty’s defense of immunity is such an affirmative defense. See, e.g., Doe v. GTE Corp., 347 F.3d 655, 657 (7th Cir.2003). Furthermore, even if Blue Buffalo were required to plead around this defense, it has sufficiently alleged that PRCG/Haggerty “designed and built” the advertising campaign challenged in the SAC.