The plaintiffs claim that bad actors misused Tile’s tracking devices to stalk them. The plaintiffs (as a class action) sued Tile for how it designed and sold the trackers. Tile invoked the arbitration clause in its TOS. This sets up a complicated analysis of TOS amendment–though the panel never directly acknowledges that this is an amendment case, not a case about formation in the first instance.

Two plaintiffs agreed to Tile’s TOS in 2021 and early 2023, respectively. Those versions of the TOS contained arbitration clauses that proved ineffective. In October 2023, Tile changed the arbitration provisions to include new language. To step up its existing users to the October 2023 arbitration provisions:

In October 2023, Tile sent to all accountholders the Oct. 2023 Notice—an email with the heading “Updated Terms of Service and Privacy Policy”—advising that Tile was updating its Terms. Sent to the email address provided by accountholders during registration, the Oct. 2023 Notice contained a blue-text and bolded hyperlink to the Oct. 2023 Terms. The email told accountholders that “[i]f you continue to use any of [Life360 and Tile’s] apps, or access our websites (other than to read the new terms) on or after November 26, 2023, you are agreeing to the [Oct. 2023 Terms].”

One plaintiff said the notification email went into the spam folder, but they saw it months later when they went affirmatively looking for it. The other plaintiff said they never saw the email. Both used the Tile service after the purported TOS amendment date. The district court held that neither plaintiff was bound by the October 2023 arbitration provisions, and the arbitration provisions in the earlier TOS versions partially failed due to unconscionability. Tile appealed that ruling, hoping to send these plaintiffs to arbitration.

The Ninth Circuit holds that Tile’s email put both plaintiffs on inquiry notice of the TOS amendment.

Transaction Context. “As Tile users, each Appellee provided an email address during account registration, and should have expected to receive relevant updates there while the account was active.” This is a pretty wild claim. Many services request email addresses during account registration, and yet the initial TOS formation fails. Also, just because I provide an email address during registration doesn’t mean that I assume TOS amendments will be sent via email. That may depend on how the TOS describes the amendment process–an angle this panel remarkably ignores completely.

Reasonable Disclosure. The court says the email disclosures were good enough:

The design and content of the Oct. 2023 Notice provided reasonably conspicuous notice of the Oct. 2023 Terms because the email’s design was “clear and legible,” and it provided the updated Terms through a link with “customary design elements denoting the existence of a hyperlink.” The subject line clearly stated that Tile was updating its Terms. And the body contained a hyperlink to the Oct. 2023 Terms in bold, blue text which contrasted against the white background. Although the email did not say specifically that the arbitration agreement would be updated, reasonable notice does not require the email to discuss every revision.

“Lack of other notices.” The court says “Tile could have done more to ensure that all its users were on inquiry notice of the Oct. 2023 Terms. Tile could, for example, have interrupted users’ next visit to the Tile App with a clickwrap pop-up notice.” The court says the absence of these other notices weighs against inquiry notice.

So, did the TOS amendment work? The court makes a remarkable doctrinal move, something I don’t recall seeing before. The court treats inquiry notice as a multi-factor test and says two factors weigh in favor of notice (transaction context and reasonable disclosures) and one weighs against (lack of other notices). In other words, the two pro-formation factors prevail over the anti-formation factor. But…when did the inquiry notice standards become a multi-factor test with these three factors? This methodology is novel (and dubious). The court might have said that even if other notification procedures would have been more efficacious, the email notice was good enough. This would have reached the same outcome without this weird doctrinal move.

[Hedging its bets, the court says “we do not hold that notice by mass email establishes inquiry notice in every case”].

Manifestations of Assent

Doe unambiguously manifested assent to the Oct. 2023 Terms by downloading the Tile App in March 2024 and using the Scan and Secure feature in attempting to locate her alleged stalker’s Tile Tracker….

Broad also unambiguously manifested assent to the Oct. 2023 Terms by using the Tile App in January 2024 and periodically opening the Tile App to check location-sharing settings—including, according to Tile’s records, in April 2024.

The court treats these users’ actions as occurring after the users had “inquiry notice.” Thus, the October 2023 TOS controls, and the court sends the case to arbitration.

* * *

Consider some of the wackiest aspects of this opinion:

• the court doesn’t distinguish between TOS formation and TOS amendment.
• the court doesn’t address what Tile’s TOS said about how the TOS could be amended. Did the TOS even authorize email amendment? The TOS terms would have substantial bearing on what a reasonable consumer might have thought. [Note: The court discusses the prior TOS’s arbitration language that said Tile couldn’t materially change the arbitration provisions “unless you expressly agree to them” but treats the October 2023 as sufficient “express agreement.”]
• the court doesn’t engage many of the precedents involving attempts to form TOSes by email, especially post-transaction emails (like this one).
• the court assumes that providing an email address during account registration means that the users should assume they will be getting TOS amendment notifications via email.
• the court didn’t address the many reasons why a TOS amendment email might never reach a user, such as the user’s email address having gone defunct or server-level blocking. If the user never received the email, does the court still think they are on inquiry notice? The court also doesn’t address the implications of the email going into a folder other than the user’s primary inbox, such as showing up in the spam folder. Are users on inquiry notice for everything in their spam folder? I wonder how often the judges carefully check their spam folder…
• the court created and applied a weird multi-factor test for inquiry notice.

FWIW, the court does acknowledge that some of the underlying issues are empirical questions, but it dodges those questions by citing Sellers, which said “there is very little empirical evidence regarding” Internet users’ expectations. If the data is hard to get, I guess we don’t need it?

So, what should we make of this opinion? Is this an example of the characteristically wild-‘n’-wooly jurisprudence in the Ninth Circuit’s non-precedential cases? Or perhaps an indicator the Ninth Circuit’s TOS formation jurisprudence is a mess and there is no logical or defensible through line from case to case?

Case Citation: Ireland-Gordy v. Tile, Inc., No. 25-403 (9th Cir. March 3, 2026). The CourtListener page.

The post Ninth Circuit Allows TOS Amendment by Email–Ireland-Gordy v. Tile appeared first on Technology & Marketing Law Blog.

asserts that Google’s spam filter failed to block abusive, spoofed, and spam emails from reaching her Gmail account. She says that reading these emails caused her emotional distress, and although she does not allege that the emails ever reached anyone else but her, she contends that they somehow damaged her reputation and disrupted her “professional communications and ongoing federal litigation.”

She sued Google for negligence, emotional distress, and something she calls “platform harm.” Yes, this is a pro se/IFP lawsuit. Unsurprisingly, it fails (multiple ways).

Section 230. The court says:

All the plaintiff’s claims are fundamentally based on her assertion that the defendant allowed certain unwanted messages through Gmail’s spam filters. Because she seeks to hold the defendant liable for its decisions related to monitoring and screening of emails, her claims “fall squarely within the exercise of a publisher’s role” and are “therefore subject to Section 230’s broad immunity.” [cite to Ynfante v. Google.] Finally, the emails in question came from an information content provider distinct from the defendant. The plaintiff does not allege that the defendant is the creator or developer of any of these emails.

Dor is suing Gmail for doing too little spam filtering. The court responds that spam filtering is the exercise of editorial discretion. This is an implicit but emphatic rejection of the must-carry obligations baked into arguments that email should be regulated like common carriage.

Then again, you can see how spam filters can’t win with  plaintiffs. Dor thinks there is too little filtering. The RNC thought Gmail did too much filtering. Spam filters could manage their liability only if, like Goldilocks, they got it just right. Otherwise, someone on one side or the other would rip them to shreds in court if the law allowed them to do so.

Prima Facie Case

The court alternatively dismisses the case for failure of the prima facie elements. Once again, Section 230 reform would not change the outcome of this case.

Negligence: “there is no indication that the defendant had a “special relationship
of custody or control” with the plaintiff. In addition, her claimed harms—emotional distress, reputational damage, and disruption of her communications—are attenuated from what the defendant is claimed to have done here, such that a reasonable person would not anticipate that the claimed harms were likely to result.” To be fair, it’s entirely foreseeable that an email address will receive crap spam. A reasonable person, however, doesn’t read the spam or, if they do, believe it to be true.

Emotional distress: “the plaintiff’s complaint alleges that the defendant failed to filter out certain emails, and those emails were distressing when she read them. The foreseeability element of both the intentional and negligent forms of emotional distress is lacking, because the defendant could not have plausibly foreseen that she would receive these emails from unknown third parties, and that she would be severely emotionally distressed by receiving them.”

“Platform Harm” (which the court interprets as a defamation claim): “The plaintiff does not allege that the defendant published a defamatory statement to a third party. Based on her complaint, the only person to receive the emails was herself.”

Implications

I wasn’t sure why this lawsuit got filed. Was it because Dor actually read her spam emails and thought they meant something? If so, this legal dilemma might be simply fixed with a short tutorial on how to manage spam emails.

I remind younger readers that spam used to be one of the top 3 Internet Law issues around a quarter-century ago. See my 2004 essay on the topic. We almost never see spam-related lawsuits nowadays. I think this reflects how improved spam filtering has ameliorated the issue to the point where spam is a minor nuisance. Notice how if this lawsuit succeeded (which it never had a chance of doing), holding spam filters liable for what they missed would make it impossible to offer spam filters at all.

Case Citation: Dor v. Google LLC, 2026 U.S. Dist. LEXIS 32957 (D. Conn. Feb. 13, 2026)

BONUS: Norton v. Turrentine, 2026 WL 687094 (N.D.Miss. March 11, 2026): “Given that Meta is an interactive service provider, the content at issue was provided by someone other than Meta, and the Plaintiff seeks to hold Meta liable for its purported exercise of a publisher’s traditional editorial functions (deciding whether to publish, withdraw, postpone, or alter content), it is clear Section 230 serves to bar the Plaintiff’s claims.”

Prior Blog Posts on Common Carriage

• Google Search Isn’t a Common Carrier–Richards v. Google
• Ninth Circuit Deletes RNC’s Lawsuit Over Gmail’s Spam Filter–RNC v. Google
• Google Search Isn’t a “Common Carrier” (DUH)–Ohio v. Google
• Court Blows Up Gmail’s Section 230 Protection, But Allegations of Biased Spam Filtering Still Fail–Republican National Committee v. Google
• Statement on the Supreme Court’s Ruling in Moody v. NetChoice
• Section 230 Protects Gmail’s Spam Filter–RNC v. Google
• Is Google’s Search Engine a “Common Carrier”? (Seriously???)–Ohio ex rel Yost v. Google
• Big Ruling for Free Speech: Most of Florida’s Social Media Censorship Law (SB 7072) Remains Enjoined–NetChoice v. Attorney General
• Texas and Its Amici Try to Justify Censorship in Their NetChoice v. Paxton Fifth Circuit Briefs
• Court Enjoins Texas’ Attempt to Censor Social Media, and the Opinion Is a Major Development in Internet Law–NetChoice v. Paxton
• Anti-Zionist Loses Lawsuit Over Social Media Account Suspensions–Martillo v. Facebook
• Texas Enacts Social Media Censorship Law to Benefit Anti-Vaxxers & Spammers
• 31 Bogus Passages from Florida’s Defense of Its Censorship Law–NetChoice v. Moody
• Florida Hits a New Censorial Low in Internet Regulation (Comments on SB 7072)
• Deconstructing Justice Thomas’ Pro-Censorship Statement in Knight First Amendment v. Trump
• Facebook Defeats Lawsuit Over Alleged ‘Shadowbanning’–De Souza Millan v. Facebook
• Are Social Media Services “State Actors” or “Common Carriers”?

The post Section 230 Preempts Lawsuit Over Unwanted Gmail Spam–Dor v. Google appeared first on Technology & Marketing Law Blog.

You may recall that the RNC claimed that Gmail’s spam filter was biased against Republican spam. The centerpiece of this claim was an academic study that the RNC intentionally misinterpreted, to the point where the authors criticized their willful misreading. Having had their case dismissed before, the RNC came back with new allegations:

“the RNC alleges that once it filed this lawsuit in October 2022, the email diversions ceased, despite the RNC sending even more emails leading up to and during the November 2022 election. Moreover, the RNC emphasizes that it targeted its emails to users that had engaged with RNC emails more recently and more frequently, and that Google’s own data showed that the RNC’s spam rate was within the limits suggested by Google.”

The lower court rejected Google’s Section 230 defense (incorrectly IMO) but dismissed the case on its prima facie elements. In a short memo opinion, the Ninth Circuit uncerimoniously dumps the case.

Common Carriage

Emailing down the Information Super Railroad Tracks. Created by ChatGPT Jan. 2026

In what seems to me to be consistently intellectual disingenuous “reasoning,” the pro-censorship crowd has an unrelenting fetish for wanting to treat emails like railroads. The court responds: “the relationship between an email sender and Google is an imperfect fit for the traditional carrier-passenger framework, and the RNC cites no authority extending California common-carrier regulations to the email context.”

Plus, the RNC isn’t a Gmail user (it sends emails to Gmail), so it can’t claim to be a disadvantaged customer of Gmail’s purported common-carriage service.

Negligent interference with prospective economic relations

RNC can’t show it had a special relationship with Google; the economic harm it suffered; or how Google caused it. Also, “imposing a duty of care would risk deterring beneficial spam filtering activity.” [This is a restatment of the Moderator’s Dilemma].

Unruh Act

The RNC lacks standing: “The RNC did not transact with Gmail, does not allege that it intended to sign up for Gmail services or that it encountered discriminatory terms, and does not allege that it was a user or prospective user of Gmail.”

17200

No injunction because the RNC alleges the problem stopped in October 2022.

* * *

A reminder: all spam filters are “biased.” That’s the whole point of filtering. The evidence that Gmail’s spam filter encoded partisan bias has never been credible. Also, it’s possible (likely?) that Republican political emails disproportionately triggered anti-spam filters because their emails resembled spam (see, e.g., this article), not because of grand partisan conspiracies.

As a result, the RNC should cut its losses and accept its defeat. More likely, they’ll seek Supreme Court cert because MAGA victimhood.

Case Citation: Republican National Committee v. Google Inc., 2026 WL 125195 (9th Cir. Jan. 16, 2026)

The post Ninth Circuit Deletes RNC’s Lawsuit Over Gmail’s Spam Filter–RNC v. Google appeared first on Technology & Marketing Law Blog.

This case involves the following TCPA restrictions (cleaned up): (1) “to make any call using an artificial or prerecorded voice to any telephone number assigned to a cellular telephone service,” or (2) “to initiate any telephone call to any residential telephone line using an artificial or prerecorded voice to deliver a message.” These restrictions depend on the statutory phrases “call,” “voice,” and “make”/”initiate” a voice call.

The Republican National Committee texted Howard’s cellphone with the following message:

As you can see, the text message included a video file that didn’t play unless the recipient pressed the play button. Howard, the plaintiff, says he never signed up for RNC’s missives. Did the RNC make/initiate a “voice” “call” by sending Ivanka’s video? The majority says no and dismisses the case.

The majority says that a “call” includes a text message. This reiterates long-standing Ninth Circuit precedent (the Satterfield case from 2009).

[In a footnote, the majority adds: “The mere fact that a phone can be set to shut off notifications for text messages does not mean that text messages do not, as a class, fall within the ordinary understanding of a potentially privacy-intruding “call.” A classic telephone call, which typically results in ringing of the phone, remains a “call” even if the particular user has his or her ringer turned off.”]

Citing a 2023 opinion, the majority says “voice” doesn’t include any silent communications.

Putting this together, the majority says the applicable provisions “only limit the use of artificial or prerecorded voices to begin a call.” The statutory concerns about privacy are implicated only when “the person called, upon being reached, is immediately and involuntarily subjected to the nuisance of listening to an artificial or prerecorded voice.” Thus, a voicemail with a prerecorded message implicates the provision, but if a live person asks the call recipient if it’s OK to play a prerecorded message, then the statutory requirements are satisfied.

The dissent calls out the majority for rewriting the statute, i.e., changing the statutory precondition of making or initiating a call into beginning the call.

As applied to Howard, the majority says:

The call at issue here—the text message and accompanying video file—included an artificial or prerecorded voice. But like a telephone call in which a live caller offers to play an artificial or prerecorded voice to the recipient, the RNC’s text message was made or initiated by its textual content and its silent inclusion of a ready-to-play video file….Because Howard’s voluntary engagement with the video file was a necessary intervening action between the RNC’s initial contact and the playing of the video’s artificial or prerecorded voice, any subsequent playing and hearing of the prerecorded voice in the video file is not included within the means by which the RNC made or initiated the “call.”

In a footnote, the majority adds: “We recognize that an individual’s phone settings could conceivably be changed so that, upon viewing a text message containing certain video files, the file would automatically begin to play. That would not change the ultimate result” because configuring that setting would constitute the recipient’s legal consent to play the file.

Implications

The TCPA Has Aged Poorly. The TCPA encoded assumptions about how the telephone system worked in the early 1990s. Obviously, the telephone system circa 2026 works very, very differently.

In the early 1990s, text messages didn’t really exist (though people used the pager system for similar functions), so the TCPA was not designed to cover text messages. Then, as courts extended the TCPA to that new medium, the courts didn’t contemplate that text messages might include multi-media components. This is why are applying a statute designed to regulate incoming telephone calls to a system for delivering video files. None of it makes sense.

In my view, every medium-exceptionalist statute ultimately collapses over time due to medium convergence. Congress misassumed that it could precisely define a telephone call that would accommodate changes in technology and social practices. As my long-standing idiom says, “if you can’t define it, you can’t regulate it.” Keep that in mind with the “social media” exceptionalist laws.

Was the TCPA the Right Tool for this Issue? What other claims could Howard have tried beyond the TCPA? The trespass to chattels claims come immediately to mind.

Political Advertising is Out-of-Control. Anti-advertising statutes routinely exclude political advertising. This makes sense for two reasons.

First, political advertising may receive greater constitutional protection than commercial advertising. Excluding political advertising is a way of bolstering the potential constitutional survivability of the statute.

Second, and I think more importantly, politicians don’t want to restrict their own power. “Do as I say, not as I do.” Today, many politicians are inveterate direct marketers, flooding their supporters (and sometimes their non-supporters) with requests for contributions. Why would they give up that stream of cash flows?

(My egoldman@gmail.com email address is often signed up for political spam by other “E. Goldman”s who are bad at typing, so I’ve seen how voracious political marketers can be).

If politicians really are concerned about marketer overgrazing, they ought to clean their own house–by statute, if necessary.

Case Citation: Howard v. Republican National Committee, 2026 WL 90273 (9th Cir. Jan. 13, 2026)

[Personal note: in 2025, my wife and I FINALLY cut our landline phone service. We now rely exclusively on our cellphones for personal phone service. I know that we’re late adopters to this approach. I don’t miss the landline at all. It was a magnet for telemarketers violating the TCPA. My cellphone number gets occasional telemarketing and text message spam, but it isn’t out of control…yet.]

The post What Does It Mean to Make a Voice Call in a Post-Telephone World?–Howard v. RNC appeared first on Technology & Marketing Law Blog.

Grecco alleges that fans uploaded 28 of its copyrighted images to their wikis. The opinion summarizes: “Between November 2023 and April 2024, MGPI sent thirty emails to Fandom’s designated agent for receiving copyright infringement complaints, copyright@wikia.com, with notifications of claimed infringement for the Asserted Images.”

Fandom requires all support requesters, including the submitters of any DMCA takedown notices, to create a support account that’s separate from their Wikia accounts; and it sends auto-replies to all emailed support requests notifying the submitter of this requirement. If the submitter doesn’t create a support account, the emails languish in a “suspension queue” (a/k/a the spam folder) for 14 days, then are deleted unread. “Fandom originally implemented the verification system in July 2023 after experiencing serious problems with high-volume spam attacks on its various support email addresses.”

Grecco refused to create a support account for several reasons, including not wanting to agree to Fandom’s TOS. As a result, its emailed DMCA notices went to Fandom’s suspension queue and then vanished. Later, Grecco directly emailed Fandom’s GC, and two days later the targeted items were gone.

At issue is Grecco’s contributory copyright infringement claim for Fandom’s prior non-removals. The opinion doesn’t address the applicability of the 512 online safe harbor. Instead, the opinion turns solely on whether Fandom had actual knowledge of the alleged infringements based on the emails that went to the suspension queue.

The judge says Fandom is off the hook:

there is no evidence that Fandom had actual knowledge of infringement of the Asserted Images prior to Mr. Grecco’s May 18, 2024 email to Mr. Cussen. In fact, there is evidence that no one at Fandom reviewed MGPI’s Takedown Requests, and therefore there is evidence that Fandom did not possess actual knowledge regarding infringement of the Asserted Images…

There is no evidence that Fandom’s verification system was a deliberate attempt to avoid acquiring knowledge about copyright infringement. In fact, the evidence shows that Fandom implemented its verification system due to significant spam attacks….Fandom may have been blind to the Takedown Requests, but there is no evidence that Fandom was willfully blind.

As a result, the judge dismissed the contributuory copyright infringement claim on summary judgment.

A month later, the judge issued a follow-on opinion. Grecco asked the judge to reconsider the ruling. None of its arguments gets any traction. For example:

MGPI states the Court’s MSJ Order determined Fandom “never legally ‘received’ Plaintiff’s DMCA Request” under 17 U.S.C. § 512(c)(3). However, the Court’s MSJ Order did not state such. See generally MSJ Order (noting the Court need not reach the issue of whether Fandom qualifies for safe harbor under the DMCA and not analyzing, or mentioning, 17 U.S.C. § 512).

The judge also turns down Fandom’s request for attorneys’ fees of over $600k, saying “MGPI’s case was not frivolous or objectively unreasonable.” Fandom argued that Grecco is a serial copyright plaintiff, citing “MGPI’s business model of pressuring and collecting revenue from copyright settlements.” You may recall that a “plaintiff is a troll” argument helped the defendant in Bell v. Kiffin, but the judge says Grecco proceeded in good faith here.

Implications

Since the rulings, both Grecco and Fandom have appealed to the Ninth Circuit. This is the kind of case that could get affirmed in a memo opinion, but at the Ninth Circuit, anything could happen.

This opinion, like so many before it, does not directly engage the obvious epistemological question: if a “machine” “knows” information but its human operators do not, does the machine’s operator nevertheless have legally recognizable “knowledge” of that information? In Internet Law, I used to highlight this question when I taught the Register.com v. Verio opinion (now sadly dropped from the casebook). In that case, Verio’s scraper robots were repeatedly “notified” of Register.com’s TOS, but did that mean Verio (the corporate legal entity) had knowledge of the terms? The opinion sidestepped that question because Verio conceded knowledge of the terms, but what would/should the outcome have been without that concession?

In this case, Fandom contested knowledge, and the judge sided with Fandom. Without expressly acknowledging it, the judge says that Fandom doesn’t “know” of the emails in its suspension queue so long as no humans review them. Of course, this isn’t the final word on the matter, and issues about what the machines know will keep arising in Internet Law for the indefinite future.

This opinion raises a second-order question: what, if any, hoops can service providers require copyright owners to jump through to submit their NOCIs/takedown notices? On the one hand, Grecco could have just postal-mailed the takedown notices to Fandom using the mailing address on Fandom’s DMCA agent registration and solved its problem earlier. On the other hand, if Fandom publicly discloses an email address to receive DMCA takedown notices–as the statute requires–it seems logical that an emailed takedown to that email address ought to be enough to satisfy the DMCA requirements. (A reminder that this opinion didn’t turn on the DMCA; it turned on the prima facie element of knowledge).

This ruling implies that services can impose some hurdles on incoming emails without dooming their legal defenses, but it leaves open obvious follow-on questions: How onerous can those hurdles be? Do they have to be established in good faith? Could a service sufficiently burden the process of sending emails such that the email option is not practically available to copyright owners? For example, if Fandom’s TOS for its support account required the copyright owner to waive all copyright claims against Fandom to submit the DMCA takedown notice, now what?

This case brought to mind the Waterman v. TikTok ruling, also from the Central District of California. I summarized the Waterman case:

the complaint completely strikes out, despite claiming that TikTok received and didn’t honor takedown notices. TikTok also prevailed without relying on the DMCA safe harbor–which, if the plaintiff’s allegations are correct, shouldn’t be available to TikTok

Sound familiar? Maybe these rulings are completely unrelated, or maybe they signal some kind of judicial fatigue over photographers’ copyright claims?

Case Citation: Michael Grecco Productions, Inc. v. Fandom, Inc., 2:24-cv-05963-MWC-BFM (C.D. Cal.). The ruling on Fandom’s motion for summary judgment (May 9, 2025). The ruling on Grecco’s request for reconsideration and Fandom’s attorneys’ fee request, 2025 WL 2271482 (July 9, 2025).

Nomenclature note: The district court judge in this case is named Judge Court, a perfect but confusing name for someone who seemed destined for the bench.

The post What’s the Difference Between Copyright Takedown Notices and Spam?–Michael Grecco v. Fandom appeared first on Technology & Marketing Law Blog.

Capital One defended on Section 230 grounds (among others). Capital One argued that the plaintiff is trying to hold Capital One liable for the text message sent by its customer to the customer’s “friend.” The plaintiff pointed out that Capital One authored and provided its customer with the exact content that the customer sent, so the content at issue wasn’t third-party content to Capital One. The district court agrees with the plaintiff.

It doesn’t change the 230 analysis that the promoting customer could have edited the content, because that didn’t happen in this situation. “Because Jensen alleges that Capital One is the sole author of the content of the text that she received, Capital One is not alleged to be merely the passive conduit of content created by others.” Groan. As I’ve discussed ad naseum on the blog, the “passive conduit” phrase is conceptually incoherent, and it’s inconsistent with Section 230’s protection for editorial decisions regarding third-party content.

The court adds: “the purpose of Section 230 immunity—to encourage Internet service providers to voluntarily monitor and edit user-generated speech in internet traffic—would not be served by protecting Capital One from liability in this case.” I disagree with the court’s characterization of Section 230’s goals, but I can see why the 230 defense vexed the court. Capital One isn’t hosting or distributing content authored by others; Capital One is trying to avoid liability for ad copy it prepared with the intent of profiting from securing new customers.

Still, the opinion sidesteps a key conceptual problem with this case. The court could have said that Capital One’s ad copy didn’t cause the legal violation asserted by the plaintiff. Capital One isn’t liable merely for providing promotional content to its customers; the customer could have been equally liable for violating the anti-spam law if they had written their own ad copy and sent it to their “friend”; and the promotional content wouldn’t have created liability if the sender had otherwise complied with the prerequisites of Washington’s anti-spam law. In other words, any anti-spam liability turns solely on the sender’s compliance efforts (or lack thereof), something Capital One can’t directly control. Viewed this way, the plaintiff seeks to hold Capital One liable for third-party “content,” i.e., its customers’ incomplete legal compliance when disseminating messages the senders had sole editorial control over.

Similar allocation-of-liability issues arose during the adware and affiliate program wars of the 2000s. See this roundup. Section 230 wasn’t a main issue in most of those litigation battles, and it doesn’t work here. Capital One may have tenable defenses on other grounds.

You may have noticed that I sometimes put the term “friends” in quotes. True friends don’t spam each other. That rule doesn’t change just because a marketing team labels its affiliate program “refer-a-friend.”

Case Citation: Jensen v. Capital One Financial Corp., 2025 WL 606194 (W.D. Wash. Feb. 25, 2025)

The post Section 230 Doesn’t Apply to “Refer-a-Friend” Text Message–Jensen v. Capital One appeared first on Technology & Marketing Law Blog.

You may recall that the RNC claimed that Gmail’s spam filter was biased against Republican spam. The centerpiece of this claim was an academic study that the RNC intentionally misinterpreted, to the point where the authors criticized their willful misreading. Having had their case dismissed before, the RNC came back with new allegations:

the RNC alleges that once it filed this lawsuit in October 2022, the email diversions ceased, despite the RNC sending even more emails leading up to and during the November 2022 election. Moreover, the RNC emphasizes that it targeted its emails to users that had engaged with RNC emails more recently and more frequently, and that Google’s own data showed that the RNC’s spam rate was within the limits suggested by Google.

They still lose. The court summarizes its conclusion:

While the RNC may be correct that Google’s alleged conduct (if proven) is “unfair” in a colloquial sense, the RNC is unable to point to any legislative policy that is implicated by the alleged conduct. Nor can it point to a sufficient harm to users of Gmail — which is the focus of the UCL — that would suggest Google’s practices are unfair. And the RNC has not shown Google’s alleged conduct has violated any other law

That’s the right outcome, but the court’s rough treatment of Section 230 this time makes me sad.

Section 230(c)(2)(A)

Defendants reluctantly rely on the Section 230(c)(2)(A) defense because of its “good faith” prerequisite. Though it succeeded last time, Section 230(c)(2)(A) now fails because Google’s allegedly changed filtering practices post-lawsuit could be evidence of Google’s intentional diversion:

while there may be technical reasons to account for the abrupt end to the months-long inboxing pattern, the timing and the lack of a clear reason for the monthly diversions makes the RNC’s allegation that Google acted without good faith in diverting the RNC’s emails to spam sufficiently plausible at this early stage of the proceedings.

In other words, it just take an unproven hypothesis to overcome Section 230(c)(2)(A)’s “good faith” standard on a motion to dismiss. Not a very helpful safe harbor.

Section 230(c)(1)

The court says: “there is no allegation that Google published or failed to remove some potentially harmful content that caused an injury leading to the RNC’s claims; rather, the challenge is to Google’s decision to restrict the availability of, or to not publish, the RNC’s emails.” The court then couples this with the repeatedly rejected argument that if Section 230(c)(1) applies to removals, Section 230(c)(2)(A) becomes superfluous (“it is subsections (c)(2)(A) and (B) that explicitly provide protection for the act of filtering, or not publishing, content provided by third parties”).

I don’t even know what to say. The court is disregarding the dozens of cases where Section 230 has applied to removal of the plaintiff’s content, so the court’s standard is emphatically wrong. The whole (c)(1)/(c)(2)(A) superfluous issue has been discussed countless times. The last time I discussed this issue, I wrote:

Section 230(c)(2)(A) fills any gaps left open by 230(c)(1). In particular, it covers suits where an ICS provider is ALSO acting as an ICP, and thus ineligible for 230(c)(1). In that circumstance, the ICS provider can still claim the statutory immunity for its filtering or removal decisions if those are exercised in good faith. Otherwise, the filtering and removal decisions are covered by 230(c)(1) and can be done however the service wants.

The court doesn’t engage with any of the discussion about this issue over many years. Sigh.

The court says Section 230(c)(2)(B) also doesn’t apply here:

If Google can show that the decisions to filter were in fact based on a user’s individualized feedback such that Google was effectively just providing the user with the means to filter the RNC’s emails, Google may then be entitled to immunity under subsection (c)(2)(B). But, taking the RNC’s allegations as true, it was Google making the filtering decisions, at least in part, not Google merely providing the technical means for filtering to its users.

The court summarizes its rejection of Section 230:

section 230 subsections (c)(1) and (c)(2)(B) do not apply in this case. And section 230(c)(2)(A) — which could potentially apply — does not bar this suit given the RNC’s allegations that Google was not operating in good faith

This is a terrible ruling for email service providers and their efforts to keep spam out of their users’ inboxes. It suggests that Section 230 does nothing for email service providers because spammers can always concoct some story suggesting bad faith filtering. Without a statutory fast lane like Section 230, spammers can tie up email service providers in lengthy and costly court proceedings based on wispy and self-serving allegations. Compare the analogous decade-long battle in Enigma v. Malwarebytes. Indeed, this same judge, in the prior ruling in this case, issued these warnings:

Permitting suits to go forward against a service provider based on the over-filtering of mass marketing emails would discourage providers from offering spam filters or significantly decrease the number of emails segregated. It would also place courts in the business of micromanaging content providers’ filtering systems in contravention of Congress’s directive that it be the provider or user that determines what is objectionable (subject to a provider acting in bad faith). This concern is exemplified by the fact that the study on which the RNC relies to show bad faith states that each of the three email systems had some sort of right- or left- leaning bias. While Google’s bias was greater than that of Yahoo or Outlook, the RNC offers no limiting principle as to how much “bias” is permissible, if any.

All good points. Too bad the court drifted away from them.

The shredding of Section 230 comes as a consequence of an RNC effort to vindicate spammers’ rights. This isn’t the first concerted Republican effort to do so; recall for example the Texas social media censorship law tried to ban email spam filters. I guess the Republicans deserve some credit for making demonstrable progress in advancing their vision for America and giving voice to an important constituency.

UCL (17200)

Despite tearing down Section 230, the RNC still loses on the prima facie elements. This is yet another example of where Section 230 reform wouldn’t change the outcome (and Section 230 isn’t inhibiting the common law’s development).

To establish a 17200 claim based on unfairness, the RNC has to find some legal principles supporting it. It can’t do so. It pointed to the Unruh Act, but the court previously held that the Unruh Act doesn’t apply to discrimination based on political affiliation.

The court then turns to the dicey subject of whether email should be treated under common carrier laws. The court again says no:

Reading email into the common carrier law would implicate significant policy and Constitutional considerations that the California Legislature has not addressed. As the Court previously discussed, “if email providers are common carriers, they would have an obligation to deliver each of the messages that were entrusted to them” including unwanted and spam emails that could be harmful and disruptive to email users and providers. And such a regulation would impose on email providers’ First Amendment rights. [cite to Moody]

The discussion of Moody is interesting because the court skipped over the part where the Moody majority (plus Justice Barrett’s concurrence) suggested that “transmitting direct messages” might be treated differently from “curating a feed.” I personally think this passage was based on underinformed stereotypes (and Joan Biskupic of CNN provides more explanation for this specific passage), but the RNC will surely question the court’s sidestepping of this part of Moody on appeal.

(Note: in Moody, Justice Barrett’s concurrence explicitly said that partisan content filtering was protected by the First Amendment. If email services are protected by the First Amendment, then the RNC’s case should be over on First Amendment grounds).

The court next turns to the UCL’s protection of consumers (“Because the RNC is not a Gmail user or Google competitor, the harms it has allegedly suffered are not properly considered under the UCL”). The court isn’t concerned about Gmail users not getting RNC spam:

Having a small number of wanted emails diverted to spam on occasion is not “substantially injurious” to Gmail users. Google is not alleged to have diverted the emails to force users to pay large sums of money to get their emails back; the users could access those emails at any time. Nor did Google realize any monetary benefit from diverting the RNC’s emails. While the practice did allegedly cause substantial monetary injury to the RNC, the Gmail users were not harmed in a similar way

(Indeed, arguably the Gmail users saved money from not pouring more money into the Trump legal defense fund).

The court summarizes:

The allegations of political discrimination, if true, are certainly concerning and may have wide and severe implications for the future of political discourse. It may even be that Google’s conduct is “unfair” in a colloquial, as opposed to a legal, sense. But it is not the role of this Court to decide these significant policy issues that must be addressed by a legislative body in the first instance

To be clear: The Texas legislature did in fact address this issue. Maybe the RNC will take up the matter there.

The court rejects the IIED claim because it’s never IIED.

* * *

The court dismissed the RNC claims with prejudice, teeing up the case for the Ninth Circuit. The opinion has several sharp edges for the Ninth Circuit to mull over, including Section 230 and the First Amendment. I can’t wait.

I imagine this case will get appealed to the Supreme Court, regardless of what happens at the Ninth Circuit. As I said in my Moody paper,

Justice Kagan once joked that the justices “are not, like, the nine greatest experts on the internet,” but they will necessarily become more Internet-savvy to review the censorial Internet laws flooding their docket.

Case Citation: Republican National Committee v. Google LLC, 2024 WL 3595538 (E.D. Cal. July 31, 2024)

The post Court Blows Up Gmail’s Section 230 Protection, But Allegations of Biased Spam Filtering Still Fail–Republican National Committee v. Google appeared first on Technology & Marketing Law Blog.

Stratics Networks offers ringless voicemail and VOIP services. The court explains that, with ringless voicemails, “users can mass deliver prerecorded messages directly to recipients’ voicemail inboxes without causing their phones to ring or giving recipients the opportunity to answer or block the call.”

[Note 1: I HATE getting voicemails. I don’t believe I’m the only person who feels this way. My outgoing voicemail message tells people to hang up and send me an email. Unlike voicemail, my email inbox is spam-filtered (unless places like Texas have successfully banned spam filters), and emails are much quicker for me to sort than voicemails. So the idea of getting unwanted and unrequested voicemails is not appealing to me.]

[Note 2: when I worked at the law firm in the 1990s, the firm’s voicemail system allowed senders to send ringless voicemails, i.e., voicemails not preceded by a phone call. I’d be sitting in my office working, and the voicemail light would magically go on without the phone ringing. I’d have to stop what I was doing, pick up the phone, log into the voicemail system, listen to the message, and if it was substantive, take written notes on the voicemail’s content. Ugh, I hated that!]

What is a “Telephone Call”?

The first question the court must resolve is whether ringless voicemails qualify as “telephone calls” for purposes of the Telemarketing Sales Rule, which prohibits deceptive or abusive telemarketing practices. The court says they are:

An interpretation of “telephone call” that does not include ringless voicemail would defy the common meaning of “telephone call.” First, people often leave voicemails when they fail to speak live with their intended recipient. Ringless voicemail is akin to leaving a voicemail when a consumer’s phone is turned off and unable to ring or be answered. Second, most consumers access their voicemail inbox via telephone. Common meaning then suggests the FTC intended to regulate ringless voicemail under the TSR regardless of whether the underlying technology uses telephone wires.

[Note: the court says that most voicemail inboxes are accessed via telephone. OK boomer. I have all of my office voicemails forwarded to my email inbox, where I listen to them. This means I receive my voicemails instantly wherever I am. I don’t use an auto-transcription feature, but that also helps sort voicemails. So listening to voicemails via telephone is so old-school.]

The court is engaging in a standard analysis of technological definitions, which frequently age poorly as the technology evolves. In the old days, a “telephone call” had certain properties. The phone ring would sonically interrupt the recipient, disrupting their concentration or waking them up from sleep. The incoming call would tie up the telephone line so other incoming and outgoing calls weren’t possible. The “call” would be synchronous, i.e., the caller and recipient talked to each other in real-time. “Ringless voicemails” share none of these attributes. In fact, about the only thing ringless voicemails share in common with old-school telemarketing is that both involve the dissemination of audio over an electronic network. So I’m not on the same wavelength as the court about the “common meaning” of a telephone call.

We can test the court’s “common meaning” by noodling what digital voice communications are NOT “telephone calls.” Is a Facetime conversation a “telephone call”? What about a Zoom meeting? If I send an audio message by text message, is that a phone call? These questions brought to mind the old Joffe case, where the held that the TCPA applied to “text calls”–say what??? (See more possibly related blog posts on the incoherence of definitions of various electronic media). Or we can flip this around–would a ringless voicemail qualify as an “email” for CAN-SPAM purposes?

As I wrote in 2005 in the Joffe post:

Regulatory efforts to carve up marketing on a per-technology basis have failed. We can’t separate telephone from email from faxes technologically, so efforts to do so legislatively are bound to lead to weird results like an anti-telemarketing law restricting sending email to a person who will receive it as email.

It’s cool that courts are still wrestling with the same question 20 years later.

Stratics’ Eligibility for Section 23o

Stratics disseminated third-party voicemails using its ringless voicemail technology. This sounds like an easy Section 230 case, no? The court does get there, but in a way that will make your head hurt. The result is a rare FTC loss on Section 230 grounds.

ICS Provider. The court says Stratics qualifies as an “access software provider” because “Stratics provides software allowing users to transmit messages to consumers’ voicemails.” The FTC argued “that because multiple users cannot access the same content, Stratics is not an interactive computer service.” I guess the FTC is taking the position that 230 only applies to public posts, not private messages? That tired argument has failed numerous times before:

Twitter is still an interactive computer service for its direct messages, which are nonpublic, because multiple users can access the site. [Cite to Fields v. Twitter]. Similarly, Stratics does not need to make content publicly available to qualify as an interactive computer service. For example, Zoom still qualifies as an interactive computer service for private video calls because Section 230 only requires that Stratics “provides or enables computer access by multiple users to a computer server.”  [Cite to In re Zoom] Finally, the CDA does not require Stratics to distribute content over the internet. For example, Twitter is still an interactive computer service even if some messages are delivered via text messages because users can access the service via the internet. [Cite to Nunes v. Twitter].

To disqualify Stratics as an access software provider, the FTC argued that it was suing over Stratics’ offline conduct. “Plaintiff argues to grant Stratics immunity would be akin to granting a newspaper immunity for publishing defamatory advertisements in hardcopy form because those advertisements were submitted via an online webportal.” If offline distribution was in fact the correct analogy, then I would agree that Section 230 doesn’t apply, as I just blogged earlier this week. However, the court doesn’t see it the FTC’s way: “It ultimately does not matter that voicemail messages are delivered to consumers’ voicemails and are not accessed via a computer. Stratics would be forced to perform the content moderation responsibilities the CDA seeks to avoid if it were held liable.” The court might have strengthened this discussion by expressly saying that Section 230 applies to cyberspace, not only “the Internet,” and that includes electronic networks primarily moving audio.

Publisher/Speaker Claim. 

Plaintiff’s Complaint alleges Stratics provided substantial assistance to sellers or telemarketers in violation of the TSR for: (1) initiating outbound prerecorded telephone calls to induce the purchase of goods or services; (2) initiating outbound telephone calls to consumers on the National DNC registry to induce the purchase of goods or services; and (3) failing to disclose the identity of the seller of the goods or services truthfully, promptly, and clearly. Each of these claims derives from Stratics’s status as a publisher or speaker

The court discusses two Section 230 workarounds, neither of which apply.

First, the court says “Section 230 does not immunize providers against content-neutral claims.” The court is referring to HomeAway, but this decontextualized statement invokes the vexing Constitutional terminology about “content-neutral” regulations. (Also, as I repeatedly remind everyone, the phrase “content neutrality” is typically an oxymoron). I think the court is using the phrase to distinguish conduct-based claims from content-based claims? If so, it should have just said that and left the “neutrality” term out of it. In any case, the court says this workaround doesn’t apply:

With respect to content neutrality, each alleged violation is premised on the content of the messages delivered using Stratics’s ringless voicemail system. While Stratics’s conduct—namely its ringless voicemail delivery system—is the basis for its alleged substantial assistance, Stratics is only liable because of the content of the messages delivered. Claims 1 and 2 allege Stratics delivered calls to induce the purchase of goods or services. If Stratics’s customers were not hawking their wares, Stratics would not be liable for providing substantial assistance. Claim 3 alleges Stratics provided substantial assistance to Stratics customers who did not appropriately disclose their identities in attempting to induce the purchase of goods or services. If Stratics’s customers had disclosed their identities, Stratics would not be liable. Accordingly, these claims are not content-neutral….

To avoid liability for providing substantial assistance, Stratics must monitor and screen the content of users’ voicemail messages. The CDA was enacted specifically to avoid imposing these moderation duties on providers.

So messy. Section 230 says websites aren’t liable for third-party content. Stratics disseminated third-party content, and the FTC wants to hold Stratics liable for that content. This isn’t complicated; whereas any judicial inquiry into “content neutrality” is going to create drama–and errors.

Second, the court says that “Section 230 permits claims that do not derive from the defendant’s status as a publisher or speaker…For example, contractual duties, supervisory duties, and tort duties, including a failure to warn, may still be brought against providers because those claims arise outside the zone of free speech and publisher interests Congress sought to protect in enacting the CDA.” That’s a pretty gross oversimplification. If those claims are being wielded to hold defendants liable for third-party content, then Section 230 applies. I think the court is trying to say that the subject matter of the third-party content is a but-for cause of the FTC’s claim; but recall that but-for prerequisite is exactly what the lower court in Neville v. Snap expressly rejected.

In any case, the court responds:

Stratics allegedly provides substantial assistance to sellers and telemarketers by virtue of distributing their content. Distribution of content is quintessential publishing activity.

It really is that simple.

Third-Party Content. “Stratics’s users generate the prerecorded messages distributed using Stratics’s platform. While Stratics developed the ringless voicemail technology at issue, that development goes to how the third-party content is distributed rather than the content itself. Therefore, third parties generate the offensive content alleged in Plaintiff’s Complaint.”

Implications. I’m sure the FTC will appeal this ruling. The FTC hates Section 230, they never want to limit their range of defendants, and they may be concerned that the court’s holding could infect the FTC’s over-interpretation of the FTC Act. On appeal, it wouldn’t surprise me if the appellate court analyzes the case differently. I still think Section 230 should apply, but the Ninth Circuit should get away from any “content neutrality” articulation of Section 230 and provide a cleaner approach to the content/conduct divide. Pending appeals in other cases will also push the Ninth Circuit to clarify the conduct/content distinction, probably before this appeal is heard.

Case Citation: U.S. v. Stratics Networks, Inc., 2024 WL 966380 (S.D. Cal. March 6, 2024)

The post “Ringless Voicemail” Vendor Wins Section 230 Defense Against FTC–US v. Stratics Networks appeared first on Technology & Marketing Law Blog.

Introduction

This lawsuit is one of the many lawsuits around the country brought by conservatives exorcising their persecution complex. Typically, these lawsuits are purely about partisanship. The lawsuits seek to work the ref, hijack other people’s resources for partisan purposes, and rile the partisan’s base. They are definitely not about the legal or policy merits. So far the judicial system has largely thwarted these partisan attacks on the legal system, but it would take only a few pro-partisan rulings (such as the Missouri v. Biden ruling, or pretty much any recent Internet Law ruling from the 5th Circuit) to radically rescramble our democracy.

* * *

In this installment of grievance litigation, the Republican National Committee claims that Gmail filtered its emails requesting donations, especially during its peak end-of-the-month fundraising appeals, because of partisan bias. The evidence for this was scant. The RNC complained that its end-of-month financial appeals weren’t as successful as it hoped. The RNC also cited the notorious North Carolina State study, which suggested that freshly minted Gmail accounts filtered conservative emails more frequently than liberal ones–despite the study authors repeatedly disclaiming that their study supported the RNC’s interpretation.

Do these data points prove Gmail’s anti-Republican bias? A few other data points contradict that inference. First, the RNC ran an A/B test where it varied the email’s outlinks in its fundraising appeals. One set of emails got through fine; another didn’t. The A/B test suggests Gmail’s filter flagged the RNC’s email coding, not its content or sender identity. Second, the Gmail team worked with the RNC for a year to help the RNC improve their email deliverability, including personally training RNC staff. The RNC thanked the Gmail team for the personal handholding by suing them.

[Also, I wonder if Google provided equal email training to the DNC. If not, Gmail apparently exhibited pro-Republican bias.]

The Opinion

Section 230(c)(2)(A)

The opinion applies Section 230(c)(2)(A), the relatively lightly litigated sibling to Section 230(c)(1). We haven’t had an email filtering case involving 230(c)(2)(A) for years, but as I wrote in 2011, “if an email service provider blocks your email, the courts aren’t going to help you out.” Thus, it’s not surprising when the court says plainly: “Plaintiff’s suit is barred because Google is entitled to immunity from suit under section 230[(c)(2)(A)] of the Communications Decency Act.”

ICS Provider. Gmail qualifies. Cite to Holomaxx v. Microsoft.

Objectionable Content. “a provider such as Google can filter spam, including marketing emails, as ‘objectionable’ material under section 230.” Cite to CAN-SPAM.

Most spam cases involve unsolicited email. In contrast, the RNC emails people who subscribed to their email lists. The court says those subscriptions don’t change the legitimacy of Gmail’s “objectionable” determination:

The fact that the RNC sent emails to individuals who requested them at some point in time does not undermine this conclusion….just because the RNC complies with the CAN-SPAM Act does not preclude that Google may reasonably consider multiple marketing emails to be “objectionable.”…Most individuals who use email are likely familiar with having engaged with an entity one time (such as by purchasing a particular product) only to have that entity send numerous other emails, many or all of which are no longer relevant or wanted. While a user may be generally able to opt out of those emails, an email provider such as Google may reasonably segregate those sorts of mass mailings

RNC sends out a significant number of emails to individuals on its list. While it may be that some, perhaps many, users specifically wanted each and every one of those emails, Google could reasonably consider these mass mailings to be objectionable, just as it can for other email senders.

Good Faith. Section 230(c)(2)(A) requires that the defendant make its filtering decision in “good faith,” something that plaintiffs challenge. In this case, the court says the RNC’s allegations about Gmail’s lack of good faith aren’t sufficient to survive a motion to dismiss–thus preventing this case from getting dragged into expensive and time-consuming discovery. The court explains:

the RNC’s allegation that Google acted in “bad faith” does not rise above the speculative level. At bottom, the RNC’s allegation is that Google diverted emails to spam at the end of the month which had been, coincidentally, a historically successful fundraising time for the RNC, and that the reasons Google gave for the low “inboxing” rate were — in the RNC’s view — not true. Plaintiff argues that the only reasonable inference for why its emails were labelled as spam is Google’s alleged political animus toward the RNC. This is pure speculation, lacking facts from which the Court could infer animus or an absence of good faith.

The North Carolina State study didn’t prove any motive on Google’s part. Plus, Google’s support for the RNC email team did improve the RNC’s email results (just not enough to satisfy RNC). The court summarizes: “the fact that Google engaged with the RNC for nearly a year and made suggestions that improved email performance is inconsistent with a lack of good faith.”

The court summarizes its policy concerns:

Permitting suits to go forward against a service provider based on the over-filtering of mass marketing emails would discourage providers from offering spam filters or significantly decrease the number of emails segregated. It would also place courts in the business of micromanaging content providers’ filtering systems in contravention of Congress’s directive that it be the provider or user that determines what is objectionable (subject to a provider acting in bad faith). This concern is exemplified by the fact that the study on which the RNC relies to show bad faith states that each of the three email systems had some sort of right- or left- leaning bias. While Google’s bias was greater than that of Yahoo or Outlook, the RNC offers no limiting principle as to how much “bias” is permissible, if any.

Amen. All spam filters exhibit multiple forms of “bias.” Allowing that to be the basis of a lawsuit creates a perpetual litigation machine.

Injunctive Relief. I plotzed when I saw the RNC argued that Section 230 only limits damages, not equitable relief. This argument has been soundly rejected for over 20 years, and I can’t believe any lawyer still thinks it might win. Note the dates of the court’s citations in rejecting this argument:

courts have rejected such a theory as it applies to liability under section 230(c)(1), and have questioned whether the theory would be viable as to section 230(c)(2). See Ben Ezra, Weinstein, & Co., Inc. v. Am. Online Inc., 206 F.3d 980, 986 (10th Cir. 2000); Smith v. Intercosmos Media Grp., Inc., No. CIV. A. 02-1964, 2002 WL 31844907, at *4–5 (E.D. La. Dec. 17, 2002); Noah v. AOL Time Warner, Inc., 261 F. Supp. 2d 532, 539–40 (E.D. Va. 2003), aff’d, No. 03-1770, 2004 WL 602711 (4th Cir. Mar. 24, 2004); Kathleen R. v. City of Livermore, 87 Cal. App. 4th 684, 698 (2001).

The cited precedents are older than some of my Internet Law students LOL. All of the cited cases predate the start of my blog in Feb. 2005. Embarrassing.

* * *

So, Gmail qualifies for Section 230(c)(2)(A). The court nevertheless also rejects the underlying claims on their lack of merit. This is yet another reminder that Section 230 reform would not change the outcome of this case.

Common Carriage. The court says: “no court, much less a court interpreting California’s common carrier law, has found an email service provider to be a common carrier. This Court declines to be the first.” It reiterates: “the RNC has not cited any authority to establish that an email provider such as Google is a common carrier, and the Court is unaware of any.”

A business may be a common carrier when it:

(1) “holds itself out to the public generally and indifferently.” Gmail satisfies this standard because anyone can register for a Gmail account. Google argued its T&Cs meant that Google wasn’t indifferent about who used it and for what purpose. The court responds: “Google prohibits certain types of email. This is no different than a mail carrier refusing to carry a dangerous or explosive device.” Is it, though? There’s a pretty big difference between speech and deadly chattel.

(2) provides services for a profit. Google satisfies this element too.

(3) “transport[s] … messages from place to place.” This does not describe an email service provider because it’s an “edge” provider, not an Internet pipe:

RNC does not allege it is a user of Gmail, but rather, alleges that its email service provider is Salesforce, and that it uses a separate email-delivery platform, Everest. As with other similar services, their email provider transforms the email into “packets” that are sent through the internet via computers on the network, and periodically reassembled and repacketized by intermediate computers on the network. Based on the address indicated in the email, the packets that constitute the email are delivered to the recipient’s email server, which are then reconstituted to form the email which is displayed the next time the recipient uses their email program. It is thus not Google that transports the messages, but rather the various computers that comprise the network. Unlike a traditional mail service, email services, like Google’s Gmail, do not “carry” messages; they receive and store messages, and make them available for retrieval by the user after the message has been shuttled through the email protocol….

services which transport information without additional processing are akin to a historical carrier service which shuttles goods and persons from point A to point B, whereas a provider of more extensive services, like Google’s Gmail, does not provide a primarily “carrier” service….

At bottom, Google does not “carry” messages. It rather receives messages from other email platforms that are carried by a decentralized computer network and displays those messages to users in the Gmail platform. It is thus not a common carrier under California Civil Code section 2168.

If you want an offline analogy, it’s why the postal service is a common carrier but the local mailbox vendor offering PO Boxes is not a common carrier. The postal service is the network, the PO Box vendor is a node at the edge of the network.

The court distinguished the abysmal Fifth Circuit NetChoice v. Paxton ruling, saying “in NetChoice II the [Texas] legislature explicitly defined social media platforms as common carriers, whereas the California legislature has not.”

Even if a legislature could overcome these considerations, it would be terrible policy to do so. The court notes the troubling policy implications of the common carriage argument:

[it] would dramatically alter the manner in which email providers conduct their business. As noted by the Plaintiff, many major email providers, including Google, have an interest in limiting spam being delivered to users…Yet, if email providers are common carriers, they would have an obligation to deliver each of the messages that were entrusted to them, as Plaintiffs themselves allege. Email providers such as Google, Yahoo, MSN and others would likely be prohibited from filtering spam or other messages and would instead be required to simply dump all emails into a user’s inbox, first come, first served. While it is true that California courts have not hesitated to interpret statutes in light of new technologies, this Court declines to accept the RNC’s invitation to interpret California’s common carrier law in such a way as to require email providers to deliver spam to the millions of Americans who use their services.

Yes, even judges hate spam.

Unruh Act. The Unruh Act is California’s flagship anti-discrimination law. It’s become a favorite of pro-censorship advocates, regardless of their partisan stripes. The court says the Unruh Act doesn’t apply to discrimination based on “political affiliations.” It also only applies to the business’ treatment of customers, and RNC wasn’t a Gmail customer.

Intentional Tortious Interference. Gmail interposes itself between the RNC and its email subscribers. However, the RNC didn’t establish a predicate wrongful act. For example, the RNC pointed to the political discrimination, which the court already held was permissible.

Negligent Tortious Interference. This legal doctrine is hard to establish without privity. Some of the factors the court considered:

• “Plaintiff alleges that its supporters have requested emails from the RNC, but this is not enough to show that Google and the Gmail users specifically contemplated the RNC’s benefit when entering into their relationship. The fact that the RNC can effectively solicit donations through Gmail is merely incidental to the relationship between Google and Gmail users.”
• The RNC concedes that the primary purpose of the spam filter is to conceal unsolicited and unwanted bulk-emailed messages as a service to its users. Its alleged negligence in not accurately filtering every email is not a moral issue… On the whole, Google’s spam filter, though in this instance imperfect, is not morally blameworthy.”
• “While no spam filter is perfect, it would not be sound policy to disincentivize the use of spam filters by imposing liability for negligent filtering. In addition, the onus of preventing the harm does not rest solely on Google. The RNC is free to solicit funds through different channels, to pay for advertising space within Gmail, or to work towards crafting emails which are not recognized as spam.”

Again, judges hate spam. The court reiterates its policy concerns:

If the Court adopted the theory of liability that Plaintiff requests, it would mean that any email provider could be liable to any of the third-party advertisers or solicitors to whom an email user may donate or purchase from (or possibly only those who make the email providers aware of their solicitation). Either way, such a large and unwieldy group of potential plaintiffs is quite unlike the “fixed, definable and contemplated group[s]” to which a duty of care has been extended. The Court finds no compelling policy justification for such a broad rule of liability.

Implications

This opinion rejected the RNC two different ways (230 and the prima facie case) in a well-constructed and thoughtful opinion. Undeterred, the plaintiffs refiled an amended complaint a couple of weeks ago.

This case reminds me of the 9th Circuit’s Enigma v. Malwarebytes decision. In that opinion, the court sided with the allegedly spyware purveyor, saying that Section 230(c)(2) didn’t apply when the defendant exercised its editorial discretion allegedly based on “anticompetitive animus.” The Ninth Circuit thus gave spyware and other threats vendors additional legal leverage to work the refs. Not only did that opinion reach a doctrinally unsound result, but it literally could make the Internet less safe.

In contrast, this opinion fully embraces the unwanted policy consequences of the RNC’s arguments. The RNC’s arguments are the same as spammers made decades ago, and historically I would have thought that adopting pro-spam positions would offend RNC contributors. I guess #MAGA is on Team Spam now. Fortunately, the judge, like most non-partisans, hates spam enough to recognize the unwanted implications of the RNC’s pro-spam arguments.

(Just a reminder that the Federal Election Commission also rejected the RNC’s complaint about Gmail filtering).

More generally, this case rejects the premise that once a consumer signs up to an email list, the email list operator has legally unrestricted rights to reach the consumer. Email isn’t purely a conversation between email senders and recipients. Those messages transmit over the email service provider’s network, and the email service must take numerous steps to manage its network, both from technical and T&S standpoints.

I would find it more amusing that the RNC got Gmail support and training, only to whine that it still wasn’t enough, except that this is how all marketers respond. Marketers always want an infinite number of customers at zero cost, and they are never satisfied.

While treating email service providers as common carriers has always been stupid, don’t forget that Texas restricted Gmail’s use of email filters in its “Social Media” censorship law. That law requires all email service providers to provide appellate rights to  spammers they filter, a non-scalable and financially ruinous obligation that would in practice force email services to turn off spam filtering entirely. The email filtering portion of Texas’ law has never been challenged in court, and the RNC could simply call in a favor from its pal Ken Paxton and weaponize this provision tomorrow (of course, Paxton’s office still has bigger problems to deal with). The Texas anti-spam filter law demonstrated just how little the Texas legislature actually cared about advancing their constituents’ well-being, but I guarantee that Texans would vote out the incumbents (or worse) if email services turned off their spam filters as the legislature effectively directed them to do. Fortunately, this ruling suggests that the Texas law would be preempted by Section 230(c)(2)(A) and may have other problems. Perhaps Texans’ inboxes are safe from Texas legislators…for now…

Case Citation: Republican National Committee v. Google, Inc., 2023 WL 5487311 (E.D. Cal. Aug. 24, 2023)

The post Section 230 Protects Gmail’s Spam Filter–RNC v. Google appeared first on Technology & Marketing Law Blog.

Both this decision and the Florida decision are major developments in Internet Law. I think this opinion is even more significant than Florida’s because:

• it is written in a broader and more persuasive way than the Florida opinion (to be fair, the Florida judge was extremely time-stressed). Its First Amendment analysis is clear and easy-to-follow.
• the Texas law was more refined than the Florida law and had fewer sharp edges (for example, it didn’t have Florida’s stupid exception for theme park owners), so it was supposed to be more resistant to constitutional challenges.
• this opinion expressly analyzes, and rejects, mandatory transparency and digital due process obligations. Given the broad-based regulatory energy being invested throughout the country to impose transparency and digital due process requirements, this is a major precedent that will pose a quandary for those regulators.
• this opinion does not rely on Section 230 (though Section 230 would have been appropriate in this case), which reduces the risk of folks freaking out about 230.

Nevertheless, this ruling isn’t the final word. We will anxiously await the Fifth Circuit’s views on this case.

The Opinion

Social Media Services Aren’t “Common Carriers”

One of most significant pro-censorship fads of the past 18 months has been claiming that social media services are “common carriers.” This judge clearly and emphatically rejects it.

The judge rightly starts by citing Halleck (a Supreme Court precedent that the pro-censorship crowd often ignores or sidesteps), saying “Social media platforms have a First Amendment right to moderate content disseminated on their platforms.” The court continues:

The Supreme Court’s holdings in Tornillo, Hurley, and PG&E, stand for the general proposition that private companies that use editorial judgment to choose whether to publish content—and, if they do publish content, use editorial judgment to choose what they want to publish—cannot be compelled by the government to publish other content.

The court then flatly rejects the “common carrier” appellation: “This Court starts from the premise that social media platforms are not common carriers.” [cite to USTA v. FCC] The court explains (emphasis added):

Unlike broadband providers and telephone companies, social media platforms “are not engaged in indiscriminate, neutral transmission of any and all users’ speech.” User-generated content on social media platforms is screened and sometimes moderated or curated. The State balks that the screening is done by an algorithm, not a person, but whatever the method, social media platforms are not mere conduits.

I love how the court decisively rejects the algorithmic exceptionalism argument. It’s never mattered whether editorial discretion is exercised by humans or machines. Either way, it’s still editorial discretion.

Because Texas rested its arguments on the common carrier theory, the court says it could stop simply with the determination that social media services aren’t common carriers. Nevertheless it continues so it can “make a determination about whether social media platforms exercise editorial discretion or occupy a purgatory between common carrier and editor.” (Purgatory is exactly where I feel like I’ve been living for these past 18 months). The court explains (emphasis added):

Social media platforms “routinely manage . . . content, allowing most, banning some, arranging content in ways intended to make it more useful or desirable for users, sometimes adding their own content.” [cite to NetChoice v. Moody] Making those decisions entails some level of editorial discretion, even if portions of those tasks are carried out by software code. While this Court acknowledges that a social media platform’s editorial discretion does not fit neatly with our 20th Century vision of a newspaper editor hand-selecting an article to publish, focusing on whether a human or AI makes those decisions is a distraction. It is indeed new and exciting—or frightening, depending on who you ask—that algorithms do some of the work that a newspaper publisher previously did, but the core question is still whether a private company exercises editorial discretion over the dissemination of content, not the exact process used….This Court is convinced that social media platforms, or at least those covered by HB 20, curate both users and content to convey a message about the type of community the platform seeks to foster and, as such, exercise editorial discretion over their platform’s content….Without editorial discretion, social media platforms could not skew their platforms ideologically, as the State accuses of them of doing.

That bolded language…whoa! As I (and many others) have said thousands of times, Internet services’ house rules differ from service to service because their audiences have different needs, and the services are appealing to different audiences. The court unreservedly acknowledges this point, recognizing that the adoption and implementation of house rules are editorial decisions.

The Level of First Amendment Scrutiny

Having established that social media services exercise editorial discretion, it’s fairly easy to conclude that strict scrutiny applies because, after all, the law by design sought to impose censorship. The court says “HB 20 prohibits social media platforms from moderating content based on ‘viewpoint,'” and the court notes some associated problems:

• “HB 20 compels social media platforms to significantly alter and distort their products”
• “HB 20 also impermissibly burdens social media platforms’ own speech”
• “the threat of lawsuits for violating Section 7 of HB 20 chills the social media platforms’ speech rights”

The court summarizes: “HB 20’s prohibitions on “censorship” and constraints on how social media platforms disseminate content violate the First Amendment.”

In a footnote, the court distinguishes Pruneyard because “the shopping mall did not engage in expression and ‘the [mall] owner did not even allege that he objected to the content of the [speech]; nor was the access right content based.'” (Please stop treating real estate cases as analogous to online publication decisions ). The court says Rumsfeld v. FAIR “has no bearing on this Court’s holding because it did not involve government restrictions on editorial functions.”

The court then turns to the law’s mandatory transparency requirements and other digital due process obligations and sets a major new precedent by saying they too trigger strict scrutiny (emphasis added):

• “Section 2’s disclosure and operational provisions are inordinately burdensome given the unfathomably large numbers of posts on these sites and apps [and by] ‘forc[ing] elements of civil society to speak when they otherwise would have refrained.” [cite to Washington Post v. McManus].
• “The consequences of noncompliance also chill the social media platforms’ speech and application of their content moderation policies and user agreements.”

In addition to the law’s viewpoint discrimination, the court says it engages in speaker- and content-based discrimination. Citing the size-based distinction in the law (and the many censorial public statements by the bill sponsor and others in government), the court says “the Legislature intended to target large social media platforms perceived as being biased against conservative views and the State’s disagreement with the social media platforms’ editorial discretion over their platforms. The evidence thus suggests that the State discriminated between social media platforms (or speakers) for reasons that do not stand up to scrutiny.” This is a reminder that size-based statutory distinctions are a double-edged sword. Without such a distinction, a digital due process requirement can impose disproportionately large costs/obligations on small publishers that inhibit speech; with such distinctions, the classifications of who qualifies for harsher regulatory treatment may not be justifiable (not that the Texas legislature really tried to justify why it set its arbitrary cutoffs where it did). For more on size-based distinctions in Internet regulation, see this article.

Some of the statutory phrases were also impermissibly vague:

• The requirement that social media services give “equal access or visibility to” content. “A social media platform is not static snapshot in time like a hard copy newspaper. It strikes the Court as nearly impossible for a social media platform—that has at least 50 million users—to determine whether any single piece of content has “equal access or visibility” versus another piece of content given the huge numbers of users and content” (emphasis added). This is a gut check on all statutory requirements of “consistent” content moderation.
• In the law’s application to any service that “consists primarily of news, sports, entertainment, or other information or content that is not user generated but is preselected by the provider,” the court says the word “primarily” is vague.
• The law authorizes the AG to enforce “potential violations” of the law. The court responds: “Subjecting social media platforms to suit for potential violations, without a qualification, reaches almost all content moderation decisions platforms might make, further chilling their First Amendment rights.”

However, some phrases are not vague. In particular, with respect to the transparency obligations, “these provisions may suffer from infirmities, [but] the Court cannot at this time find them unconstitutionally vague on their face.”

Heightened Scrutiny

Although the court concludes that strict scrutiny applies, it says the law would not survive intermediate scrutiny either:

• “the State provides no convincing support for recognizing a governmental interest in the free and unobstructed use of common carriers’ information conduits” [cite to Miami Herald v. Tornillo, and the court recognizes that the state’s argument is a word salad too]
• “The State’s second interest—preventing ‘discrimination’ by social media platforms—has been rejected by the Supreme Court. Even given a state’s general interest in anti-discrimination laws, ‘forbidding acts of discrimination’ is ‘a decidedly fatal objective’ for the First Amendment’s ‘free speech commands.'” [cite to Hurley]
• In a footnote, the court dismisses the applicability of the confusing Turner cable precedent because it ultimately sought to preserve the integrity of the broadcasting scheme. The court says: “The analysis applied to the regulation of broadcast television has no bearing on the analysis of Internet First Amendment protections.” [cite to Reno v. ACLU, another major Supreme Court precedent that the pro-censorship forces often ignore]
• “HB 20 is not narrowly tailored. Sections 2 and 7 contain broad provisions with far-reaching, serious consequences…the [NetChoice v. Moody] court colorfully described [the Florida social media censorship law] as ‘an instance of burning the house to roast a pig.’ This Court could not do better in describing HB 20.” [As I said in my NetChoice v. Moody post, the “meme of porcine incineration and residential arson…runs throughout First Amendment jurisprudence.”]

If the law can’t survive intermediate scrutiny, then the state must find a way to qualify the law for rational basis scrutiny. There’s really no way it can. As a result, the court’s rejection on intermediate scrutiny grounds is another major precedent.

Other Issues

The law’s lengthy severability provisions (seriously, they ran for several pages) don’t help the state:

Both sections are replete with constitutional defects, including unconstitutional content- and speaker-based infringement on editorial discretion and onerously burdensome disclosure and operational requirements. Like the Florida statute, “[t]here is nothing that could be severed and survive.”

On the remaining elements of a preliminary injunction, the court adds (emphasis added): “HB 20 prohibits virtually all content moderation, the very tool that social medial platforms employ to make their platforms safe, useful, and enjoyable for users….In addition, social media platforms would lose users and advertisers, resulting in irreparable injury….content moderation and curation will benefit users and the public by reducing harmful content and providing a safe, useful service.” The bolded language is crucial because it shows the judge understood the policy stakes of this case. House rules are the sine qua non of social media. Efforts to strip away house rules essentially eliminate the services’ raison d’etre. For more on the policy considerations, see this article.

Implications

This is a major ruling. Even though it will eventually be supplanted by the Fifth Circuit’s decision, for now it casts a huge shadow over many of the pro-censorial regulatory efforts being pursued around the country. Reading this opinion, I felt a little like how I felt reading the initial district court opinion enjoining the ACLU v. Reno CDA litigation over 25 years ago. That too was a quickly issued but important ruling protecting the Internet, where a contrary result would have potentially permanently harmed the Internet’s development. So too with this opinion.

The court explicitly grounded its decision only in the First Amendment and specifically sidestepped the Section 230 and dormant Commerce Clause challenges. The court’s First Amendment-only ruling has several beneficial effects:

• It doesn’t give the appellate court any room to do something stupid about Section 230, such as question Section 230’s constitutionality or find new exceptions to Section 230.
• It preserves the court’s option to strike down the law on other grounds if the appellate court reverses it.
• It takes the wind out of the sails of critics who falsely claim that Section 230, and not the First Amendment, is the reason why state legislatures can’t regulate the crap out of social media. As many of us have been saying for a long time, Section 230 reform may not change the substantive outcomes with respect to laws like this because the First Amendment would provide a more cumbersome route to the same substantive result. Instead, the reform would just mess up Section 230 for other use cases.

When I blogged the Florida law, I invoked the meme of a barking dog that catches the car, as an analogy to what happens when a state legislature passes a #MAGA messaging bill that was unmitigated censorship. When that happens, courts should easily strike it down. While we should celebrate the judicial branch as the strongest bastion of liberty left in our country, we should never stop condemning legislators who waste their time and their taxpayers’ money passing an obviously unconstitutional law. Worse, even if the law survived, it represents the worst kind of policy because it would accomplish none of its purported goals. As Jess Miers and I explain here, the law would actually be counterproductive to every goal it claims it seeks to advance. Texas residents, you deserve better governance.

Although this ruling strikes down much of HB20’s payload, the plaintiffs didn’t challenge HB20’s must-carry-spam provision–the worst possible policy that a legislature could pursue. As a result, AG Paxton can now enforce that provision at any time. I wonder if this decisive ruling will dissuade him from doing so. It’s possible the pro-spam provision will become one of those statutes that’s technically on the books but everyone ignores. I still think someone should preemptively challenge it rather than leave it hanging.

For more on this opinion, see the Techdirt coverage.

This is an amazingly well-crafted and savvy opinion produced on a very speedy timeline. For his remarkable combination of speed and accuracy, I’m awarding the Technology & Marketing Law Blog Judge-of-the-Day award to Judge Robert L. Pitman.

Case citation: NetChoice, Inc. v. Paxton, 2021 WL 5755120 (W.D. Tex. Dec. 1, 2021)

Case library (see also NetChoice’s library and the Court Listener page):

• District court stays the proceedings pending appeal.
• District court opinion enjoining the law. My blog post.
• Plaintiffs’ reply brief.
• Amicus brief in support of Texas from Babylon Bee et al
• Amicus briefs in support of NetChoice/CCIA preliminary injunction motion
  • Chamber of Progress et al
  • EFF
  • RCFP et al
  • TechFreedom
• Texas response.
  • Adam Candeub “expert” witness report. Motion to strike this report.
• Preliminary injunction motion. My blog post.
  • NetChoice Declaration
  • CCIA Declaration
  • Declaration from YouTube
  • Declaration from Facebook
  • Declaration from LGBT Technology Institute
  • Declaration from Stop Child Predators
  • Declaration from Technology Network
• NetChoice/CCIA complaint.
• Text of HB 20. My blog post.

The post Court Enjoins Texas’ Attempt to Censor Social Media, and the Opinion Is a Major Development in Internet Law–NetChoice v. Paxton appeared first on Technology & Marketing Law Blog.