I. BACKGROUND

In February 2019, Murphy filed a complaint against Twitter, Inc. and Twitter International Company (Twitter), asserting causes of action for breach of contract, promissory estoppel, and violation of Business and Professions Code section 17200, the unfair competition law (UCL). Twitter operates an Internet communications platform that allows its users to post short messages, called “tweets,” as well as photos and short videos. Hundreds of millions of active users use Twitter to communicate, share views, and discuss issues of public interest. Twitter users can “follow” other users and thereby choose whose tweets they want to see. Meghan Murphy is a freelance journalist and writer who writes primarily on feminist issues from both a socialist and feminist perspective. Murphy is also the founder and editor of Feminist Current, a feminist blog and podcast. Murphy joined Twitter in April 2011, and used it “to discuss newsworthy events and public issues, share articles, podcasts and videos, promote and support her writing, journalism and public speaking activities,

2 and communicate with her followers.” At the time her account was permanently suspended, Murphy had approximately 25,000 followers. Twitter had also given Murphy a blue “verification badge,” which “ ‘lets people know that an account of public interest is authentic.’ ” According to her complaint, Murphy “writes primarily on feminist issues, including the Me Too movement, the sex industry, sex education, third-wave feminism, and gender identity politics.” In her work, Murphy argues “that there is a difference between acknowledging that transgender women see themselves as female and counting them as women in a legal or social sense.” She “object[s] to the notion that one’s gender is purely a matter of personal preference.” Beginning in January 2018, Murphy posted a series of tweets about Hailey Heartless, a prominent public figure who had been chosen to speak at the Vancouver Women’s March in 2018. According to the march organizers, Heartless “ ‘self identifies as a transsexual professional dominatrix’ ” and “ ‘has over ten years of activist experience in LGBTQ, feminist, sex positive, sex worker and labour communities.’ ” Heartless’s legal name is Lisa Kreut. Kreut had identified as a male until approximately three years earlier. At the 2016 British Columbia Federation of Labour (BCFED) Conference, Kreut had helped organize a successful effort to prohibit BCFED and its affiliated unions from funding the Vancouver Rape Relief and Women’s Shelter, on the ground that it limited its services to biological females. Murphy was “intensely critical of the effort to defund the Women’s Shelter.” On January 11, 2018, Murphy tweeted: “For the record, this ‘dominatrix’ was also one of those behind the push to get @bcfed to boycott and defund Vancouver Rape Relief, Canada’s longest standing rape crisis

3 center. He is ACTIVELY working to take away women’s services and harm the feminist movement.” In May 2018, Murphy again tweeted about Kreut after Kreut and others signed an open letter to organizers of the Vancouver Crossroads conference. The letter, which was posted to a website Kreut helped create, urged conference organizers to remove a local poverty activist from a panel discussion on urban renewal because she was “ ‘a well-documented Trans Exclusionary Radical Feminist (TERF) and Sex Worker Exclusionary Radical Feminist (SWERF), and is known in the community to promote this ideology.’ ” Murphy alleges the letter made clear that it was also targeted at her, and that the letter signatories were “urging that she never again be allowed to speak in public either.” In response to the open letter, Murphy tweeted: “Lisa Kreut and another trans-identified male/misogynist created a website in order to libel a local woc activist, and published a letter demanding she be removed from a panel scheduled as part of this conference . . . . The organizers caved immediately.” A second tweet posted moments later said: “The ‘evidence’ provided to claim the activist should be removed is almost entirely to do with her activism against the sex trade, then literally a few retweets and ‘likes’ from feminists these men don’t like. Seven people signed the thing. It’s ridiculous.” After Murphy’s May tweets, Kreut contacted SheKnows Publishing Network, the company that arranges advertising for Murphy’s blog, Feminist Current, to complain about Murphy’s writing. SheKnows responded in July 2018 by pulling all advertising from Feminist Current and terminating its relationship with the site. On August 30, Murphy wrote three more tweets about Kreut: “ ‘Aaaand look who publicly admitted to going after @feministcurrent’s ad revenue in an attempt to shut us down,

4 and is now offering tips to other men in order to go after @MumsnetTowers’ ” “This is Lisa Kreut, @lispinglisa, the male BDSMer who was given a platform to promote prostitution at the Vancouver Women’s March this year, who led efforts to defund Vancouver Rape Relief & Women’s Shelter at BCFED 2016 . . . .” “So @BlogHer pulled revenue from a feminist site because a white man who spends his energy promoting the sex trade as empowering for women and targeting/trying to silence/defund women’s shelters, female activists, and feminist media told them to.” The same day, Twitter locked Murphy’s account for the first time. Twitter claimed that four of Murphy’s tweets, the tweet from January 11 and the three tweets from August 30, “[v]iolat[ed] our rules against hateful conduct.” Twitter required Murphy to delete the tweets before she could regain access to her account. The next day, Murphy tweeted: “Hi @Twitter, I’m a journalist. Am I no longer allowed to report facts on your platform?” Twitter required Murphy to delete that tweet as well on the ground it violated Twitter’s “Hateful Conduct Policy” (Hateful Conduct Policy). Twitter also suspended Murphy’s account for 12 hours. Murphy appealed the suspension, but received no response. On November 15, Murphy’s account was locked again. Twitter required her to remove a tweet from October 11 that stated, “Men aren’t women,” and a tweet from October 15 that asked, “How are transwomen not men? What is the difference between a man and a transwoman?” Twitter told Murphy the tweets violated its Hateful Conduct Policy. The same day, Murphy tweeted: “This is fucking bullshit @twitter. I’m not allowed to say that men aren’t women or ask questions about the notion of transgenderism at all anymore? That a multi billion dollar company is

5 censoring BASIC FACTS and silencing people who ask questions about this dogma is INSANE . . . .” Four days later, on November 19, Twitter locked Murphy out of her account and required her to delete her tweet from November 15 in which she criticized Twitter’s actions. Twitter did not identify any rule or policy the November 15 tweet violated. On November 20, Murphy was once again locked out of her account, and required to delete her two tweets from May 2018 about Lisa Kreut. On November 23, Twitter sent Murphy a private e-mail stating that she was being permanently suspended based on a November 8 tweet in which Murphy wrote, “ ‘Yeeeah it’s him’ ” over an embedded image of a Google review of a waxing salon posted by “Jonathan Y.” (hereafter J.Y.).1 J.Y. had filed 16 different human rights complaints under the alias J.Y. against female estheticians across Canada for refusing to perform Brazilian waxes on J.Y. because J.Y. has male genitalia. On November 8, Murphy posted on Twitter, citing J.Y.’s Twitter handle, @trustednerd: “ ‘Is it true that the man responsible for trying to extort money from estheticians who refuse to give him a brazilian bikini wax is @trustednerd? Why tf is the media/court protecting this guy’s identity either way? The women he targeted don’t get that luxury.’ ” Murphy then tweeted: “ ‘This is also, it should be pointed out, a key problem with allowing men to ID as female, change their names, IDs etc. They can leave behind these kinds of pasts (and likely continue to

1Murphy’s complaint refers to J.Y. as “Jonathan [Y.].” Twitter, in its respondent’s brief, states that J.Y. is a transgender woman, Jessica Y. Though Twitter does not attribute that fact to any portion of the record, we note that Murphy seeks judicial notice of a decision from a Canadian tribunal which states “Jessica [Y.] is a transgender woman,” and refers to J.Y. with the pronouns “her” and “she.”

6 predate on women and girls, where that abuse will be reported as perpetrated by a “woman”).’ ” Twitter told Murphy the tweet violated its Hateful Conduct Policy. As adopted in 2015, Twitter’s Hateful Conduct Policy stated: “Hateful conduct: You may not promote violence against or directly attack or threaten other people on the basis of race, ethnicity, national origin, sexual orientation, gender, gender identity, religious affiliation, age, disability, or serious disease. We also do not allow accounts whose primary purpose is inciting harm towards others on the basis of these categories.” Murphy alleges Twitter amended the policy in December 2017, to add a preface stating: “Freedom of expression means little if voices are silenced because people are afraid to speak up. We do not tolerate behavior that harasses, intimidates, or uses fear to silence another person’s voice. If you see something on Twitter that violates these rules, please report it to us.” The amended policy also offered specific examples of harassing behavior Twitter does not tolerate and added a section on “How enforcement works,” which emphasized, “Context matters. [¶] . . . Some Tweets may seem to be abusive when viewed in isolation, but may not be when viewed in the context of a larger conversation. While we accept reports of violations from anyone, sometimes we also need to hear directly from the target to ensure that we have proper context.” Murphy alleges that in late October 2018, Twitter made “sweeping changes” to its Hateful Conduct Policy, “nearly tripling the policy in length,” and adding a provision that prohibited “targeting individuals with repeated slurs, tropes or other content that intends to dehumanize, degrade or reinforce negative or harmful stereotypes about a protected category. This includes targeted misgendering or deadnaming of transgender individuals.”

7 Murphy alleges Twitter instituted the new policy without providing the 30- day notice required by its own terms of service and retroactively enforced its new terms against her. Murphy also contends the new Hateful Conduct Policy is “viewpoint discriminatory on its face” because it “forbids expression of the viewpoints that 1) whether an individual is a man or a woman is determined by their sex at birth and 2) an individual’s gender is not simply a matter of personal preference,” viewpoints she alleges are “widely-held” and “shared by a majority of the American public.” She alleges the new policy contradicted “repeated promises and representations” by Twitter “that it would not ban users based on their political philosophies or viewpoints or promulgate policies barring users from expressing certain philosophies, or viewpoints.” She further alleges enforcement of the “ ‘misgendering’ ” policy requires Twitter to “engage in active content monitoring and censorship,” which its rules previously said it would not do. Twitter also amended its terms of service several times between 2012 and 2017, with respect to its right to suspend or terminate accounts and remove or refuse to distribute content provided by users. On May 18, 2015, it amended the terms of service to state, among other things: “We may suspend or terminate your accounts or cease providing you with all or part of the Services at any time for any or no reason, including, but not limited to, if we reasonably believe: (i) you have violated these Terms or the Twitter Rules . . . .” That provision was still operative in 2018, when Murphy filed her lawsuit. On October 2, 2017, Twitter further amended the terms of service to state: “We may also remove or refuse to distribute any Content on the Services, suspend or terminate users, and reclaim usernames without liability to you.” Twitter’s terms of service also state users will be notified “30 days in advance of making effective changes to [the terms of service] that

8 impact the rights or obligations of any party to these Terms,” and promises users that changes “will not be retroactive.” Murphy’s first cause of action for breach of contract alleges Twitter breached the express contractual terms of its user agreement by failing to provide notice of the changes to the Hateful Conduct Policy, including the “new ‘misgendering’ provision,” before enacting it, and by enforcing the changes against Murphy retroactively. Murphy also contends Twitter violated its user agreement and “the duty of good faith and fair dealing implicit within it” because it “targeted her for permanent suspension despite the fact that she never violated any of the Terms of Service, Rules, or incorporated policies.” She further asserts the portions of the terms of service purporting to give Twitter the right to suspend an account “at any time for any or no reason” and “without liability to you” are procedurally and substantively unconscionable. Her second cause of action for promissory estoppel alleges Twitter violated several clear and unambiguous promises in the user agreement, on its website, and in public statements, including the promise to not monitor or censor content, the promise to notify users of changes 30 days before they are made, the promise to not apply changes retroactively, a promise to reserve “account-level” actions, such as permanent suspensions, for repeated or egregious violations, and promises to treat everyone the same and not consider “ ‘political viewpoints, perspectives, or party affiliation in any of [Twitter’s] policies or enforcement decisions, period.’ ” Murphy alleges she and other users foreseeably and reasonably relied on the promises to their detriment, and that they have lost “valuable economic interests in access to their Twitter account[s] and their followers forever.”

9 Murphy’s third cause of action for violation of the UCL alleges Twitter committed an unfair business practice by inserting the alleged unconscionable provisions allowing it to suspend or ban accounts “at any time for any or no reason” and “without liability to you” into its terms of service, and that its practices are “fraudulent” within the meaning of the UCL because Twitter falsely “held itself out to be a free speech platform” and promised not to actively monitor or censor user content. In her complaint, Murphy seeks injunctive relief on behalf of herself and similarly situated individuals, requiring, among other things, that Twitter (1) cease and desist enforcing its “unannounced and viewpoint discriminatory ‘misgendering’ rule”; (2) restore accounts it had suspended or banned pursuant to the misgendering policy; (3) remove “unconscionable provisions in its terms of service purporting to give Twitter the right to suspend or ban an account ‘at any time for any or no reason’ and ‘without liability to you’ ”; and (4) issue a “full and frank correction of its false and misleading advertising and representations to the general public that it does not censor user content except in narrowly-defined, viewpoint-neutral circumstances . . . .” Murphy also requests a declaratory judgment that Twitter has breached its contractual agreements with Murphy and similarly situated users by taking the actions alleged in the complaint. Twitter filed a demurrer and special motion to strike under the anti- SLAPP law (Code Civ. Proc., § 425.16).2 Twitter argued that Murphy’s claims were barred by the immunity provided in section 230(c)(1) and the First Amendment to the United States Constitution. Twitter also asserted

2 “ ‘SLAPP’ is an acronym for ‘strategic lawsuit against public participation.’ ” (Baral v. Schnitt (2016) 1 Cal.5th 376, 381, fn. 1.)

10 Murphy had failed to state a claim as to each of her causes of action under California law. Murphy filed an opposition and Twitter filed a reply.3 After hearing oral argument, the trial court denied Twitter’s anti- SLAPP motion based on the “public interest” exception of Code of Civil Procedure section 425.17, subdivision (b). The court sustained Twitter’s demurrer without leave to amend based on section 230(c)(1) and thereafter entered a judgment of dismissal.

II. DISCUSSION

A. Communications Decency Act Section 230 is part of the Communications Decency Act of 1996. As our Supreme Court has explained, “Congress enacted section 230 ‘for two basic policy reasons: to promote the free exchange of information and ideas over the Internet and to encourage voluntary monitoring for offensive or obscene material.’ ” (Hassell v. Bird (2018) 5 Cal.5th 522, 534 (Hassell).) The statute contains express findings and policy declarations recognizing the rapid growth of the Internet, the beneficial effect of minimal government regulation on its expansion, and the twin policy goals of “promot[ing] the continued development of the Internet and other interactive computer services” and “preserv[ing] the vibrant and competitive free market that presently exists for the Internet and other interactive computer services, unfettered by Federal or State regulation.” (§ 230(a), (b).) The law was enacted in part in response to an unpublished New York trial court decision, Stratton Oakmont, Inc. v. Prodigy Services Co. (N.Y.Sup.Ct. 1995) 23 Media L.Rep. 1794 [1995 WL 323710] (Stratton Oakmont), which held that because an operator of

3Twitter apparently filed a reply brief in support of its demurrer, but it was not included in the record on appeal. The only reply brief in the record on appeal was Twitter’s reply in support of its special motion to strike under Code of Civil Procedure section 425.16.

11 Internet bulletin boards had taken an active role in monitoring and editing the content posted by third parties to the bulletin boards, it could be regarded as the “publisher” of material posted on them and held liable for defamation. (Hassell, at p. 534; Zeran v. America Online, Inc. (4th Cir. 1997) 129 F.3d 327, 331 (Zeran).) Section 230(c)(1), which is captioned “Treatment of publisher or speaker,” states: “No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider.” As relevant here, the statute also expressly preempts any state law claims inconsistent with that provision: “No cause of action may be brought and no liability may be imposed under any State or local law that is inconsistent with this section.” (§ 230(e)(3).) Read together these two provisions “protect from liability (1) a provider or user of an interactive computer service (2) whom a plaintiff seeks to treat, under a state law cause of action, as a publisher or speaker (3) of information provided by another information content provider.” (Barnes v. Yahoo!, Inc. (9th Cir. 2009) 570 F.3d 1096, 1100–1101, fn. omitted (Barnes); Delfino v. Agilent Technologies, Inc. (2006) 145 Cal.App.4th 790, 804–805 (Delfino).) Two California Supreme Court cases, Hassell, supra, 5 Cal.5th 522 and Barrett v. Rosenthal (2006) 40 Cal.4th 33 (Barrett), have addressed immunity under section 230, discussing at length statutory interpretation and judicial construction of the statute.4 In both cases, our high court concluded

4 Although Hassell was a plurality opinion, Justice Kruger, concurring in the judgment, agreed with the plurality’s analysis of section 230. (Hassell, supra, 5 Cal.5th at pp. 548, 558 (conc. opn. of Kruger, J.) [“section 230 immunity applies to an effort to bring a cause of action or impose civil liability on a computer service provider that derives from its status as a publisher or speaker of third party content”].)

12 section 230 is to be construed broadly in favor of immunity. (Hassell, at p. 544 [“broad scope of section 230 immunity” is underscored by “inclusive language” of § 230(e)(3), which, “read in connection with section 230(c)(1) and the rest of section 230, conveys an intent to shield Internet intermediaries from the burdens associated with defending against state law claims that treat them as the publisher or speaker of third party content, and from compelled compliance with demands for relief that, when viewed in the context of a plaintiff’s allegations, similarly assign them the legal role and responsibilities of a publisher qua publisher”]; Barrett, at p. 39 [immunity provisions within § 230 “have been widely and consistently interpreted to confer broad immunity”].) California’s appellate courts and federal courts have also generally interpreted section 230 to confer broad immunity on interactive computer services. (See Doe II v. MySpace Inc. (2009) 175 Cal.App.4th 561, 572 [concluding a “general consensus to interpret section 230 immunity broadly” could be derived from California and federal court cases]; Delfino, supra, 145 Cal.App.4th at p. 804; accord, Doe v. MySpace Inc. (5th Cir. 2008) 528 F.3d 413, 418; Carafano v. Metrospalsh.com, Inc. (9th Cir. 2003) 339 F.3d 1119, 1123 [“reviewing courts have treated § 230(c) immunity as quite robust”]; but see Barnes, supra, 570 F.3d at p. 1100 [text of § 230(c) “appears clear that neither this subsection nor any other declares a general immunity from liability deriving from third-party content”].) Murphy’s claims against Twitter satisfy all three conditions for immunity under section 230(c)(1). First, an “ ‘interactive computer service’ ” is “any information service, system, or access software provider that provides or enables computer access by multiple users to a computer server.” (§ 230(f)(2).) Twitter meets that description, as Murphy concedes. (See

13 American Freedom Defense Initiative v. Lynch (D.D.C. 2016) 217 F.Supp.3d 100, 104 [Twitter is an interactive computer service under the CDA]; Fields v. Twitter, Inc. (N.D.Cal. 2016) 217 F.Supp.3d 1116, 1121, affd. 881 F.3d 739 (9th Cir. 2018); Brittain v. Twitter, Inc. (N.D.Cal. Jun. 10, 2019, No. 19-cv- 00114-YGR) 2019 WL 2423375 at p. *2.) Second, Murphy’s claims all seek to hold Twitter liable for requiring her to remove tweets and suspending her Twitter account and those of other users. Twitter’s refusal to allow certain content on its platform, however, is typical publisher conduct protected by section 230. “ ‘[Section] 230 precludes courts from entertaining claims that would place a computer service provider in a publisher’s role. Thus, lawsuits seeking to hold a service provider liable for its exercise of a publisher’s traditional editorial functions—such as deciding whether to publish, withdraw, postpone or alter content—are barred.’ ” (Barrett, supra, 40 Cal.4th at p. 43, quoting Zeran, supra, 129 F.3d at p. 330; Fair Housing Council of San Fernando Valley v. Roommates.com, LLC (9th Cir. 2008) 521 F.3d 1157, 1170–1171 [“any activity that can be boiled down to deciding whether to exclude material that third parties seek to post online is perforce immune under section 230”].) The third prong is also satisfied here, because Murphy’s claims all concern Twitter’s removal of or refusal to publish “information provided by another information content provider.” An “ ‘information content provider’ ” is defined as “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.” (§ 230(f)(3).) All of the content that Murphy claims Twitter required her or others to remove and is wrongfully censoring was created and posted by Murphy and others, not Twitter. (See, e.g., Cross v. Facebook, Inc. (2017) 14 Cal.App.5th 190, 207 (Cross); Federal

14 Agency of News LLC v. Facebook, Inc. (N.D.Cal. 2020) 432 F.Supp.3d 1107, 1118 (Federal Agency of News) [complaint admitted Federal Agency of News LLC was the source of content Facebook, Inc. (Facebook) removed].) Murphy takes issue with both the second and third prongs of the section 230 test as they relate to her claims. She contends section 230(c)(1) cannot apply in this case because the “only information at issue is Twitter’s own promises,” not “ ‘information provided by another content provider,’ ” and because she seeks to treat Twitter not as a publisher of information provided by others, but as a promisor or party to a contract. Murphy relies on Barnes, supra, 570 F.3d at page 1107 and Demetriades v. Yelp, Inc. (2014) 228 Cal.App.4th 294, 313 (Demetriades) to argue that “[c]ourts routinely have held that Section 230 permits contract, promissory estoppel and consumer fraud claims” such as hers. In assessing whether a claim treats a provider as a publisher or speaker of user-generated content, however, courts focus not on the name of the cause of action, but whether the plaintiff’s claim requires the court to treat the defendant as the publisher or speaker of information created by another. (Barnes, supra, 570 F.3d at pp. 1101–1102; Cross, supra, 14 Cal.App.5th at p. 207.) This test prevents plaintiffs from avoiding the broad immunity of section 230 through the “ ‘ “creative” pleading’ of barred claims” or using “litigation strategy . . . to accomplish indirectly what Congress has clearly forbidden them to achieve directly.” (Hassell, supra, 5 Cal.5th at pp. 542, 541.) Courts have routinely rejected a wide variety of civil claims like Murphy’s that seek to hold interactive computer services liable for removing or blocking content or suspending or deleting accounts (or failing to do so) on

15 the grounds they are barred by the CDA.5 (See, e.g., Doe II v. MySpace Inc., supra, 175 Cal.App.4th at p. 573 [§ 230 immunity barred tort claims based on social networking website’s decisions whether “to restrict or make available” minors’ profiles]; Gentry v. eBay, Inc. (2002) 99 Cal.App.4th 816, 834–836 [rejecting negligence and UCL claims under § 230 and California law based on auction website’s compilation of and failure to withdraw allegedly false and misleading information provided by individual sellers on its website]; Wilson v. Twitter (S.D.W.Va. May 1, 2020, No. 3:20-cv-00054) 2020 WL 3410349 at pp. *1, *12 [plaintiff’s claims seeking to hold Twitter liable for deleting posts and suspending account based on hateful conduct policy barred by § 230(c)(1)], report and recommendation adopted, Wilson v. Twitter (S.D.W.Va. June 16, 2020, No. 3:20-cv-00054) 2020 WL 3256820; Domen v. Vimeo, Inc. (S.D.N.Y. 2020) 433 F.Supp.3d 592, 602–603 [defendant entitled to immunity under § 230(c)(1) because plaintiffs sought to treat defendant as a “ ‘publisher’ ” for deleting plaintiffs’ content on its website]; Ebeid v. Facebook, Inc. (N.D.Cal. May 9, 2019, No. 18-cv-07030-PJH) 2019 WL 2059662 at pp. *3–*5 [removal of plaintiff’s Facebook posts and restrictions on his use of his account constituted “publisher activity” protected by § 230];

5 Courts have found immunity for interactive computer services under section 230 regardless of whether the provider is alleged to have improperly removed objectionable content or failed to remove such content. “[N]o logical distinction can be drawn between a defendant who actively selects information for publication and one who screens submitted material, removing offensive content. ‘. . . the difference is one of method or degree, not substance.’ ” (Barrett, supra, 40 Cal.4th at p. 62; see Barnes, supra, 570 F.3d at p. 1102, fn. 8 [it is immaterial whether service provider’s exercise of publisher’s traditional editorial functions “comes in the form of deciding what to publish in the first place or what to remove among the published material”]; Fyk v. Facebook, Inc. (9th Cir. 2020) 808 Fed.Appx. 597, 597, fn. 2.)

16 Mezey v. Twitter, Inc. (S.D.Fla. Jul. 19, 2018, No. 1:18-cv-21069-KMM) 2018 WL 5306769, at pp. *1–*2 [lawsuit alleging Twitter “unlawfully suspended [plaintiff’s] Twitter account” dismissed on grounds of § 230(c)(1) immunity]; Fields v. Twitter, Inc. (N.D.Cal. 2016) 217 F.Supp.3d 1116, 1123–1125, affd. 881 F.3d 739 (9th Cir. 2018) [holding provision of Twitter accounts to alleged terrorists was publishing activity immunized by § 230]; Sikhs for Justice “SFJ”, Inc. v. Facebook, Inc. (N.D.Cal. 2015) 144 F.Supp.3d 1088, 1092–1093 [rejecting claim by human rights organization against Facebook, Inc. for blocking access to its website for discriminatory reasons at the request of government of India]; Riggs v. MySpace, Inc. (9th Cir. 2011) 444 Fed.Appx. 986, 987 [district court properly dismissed claims arising from service provider’s decisions to delete plaintiff’s user profiles].) While Murphy is correct that some courts have rejected the application of section 230 immunity to certain breach of contract and promissory estoppel claims, many others have concluded such claims were barred because the plaintiff’s cause of action sought to treat the defendant as a publisher or speaker of user generated content. (See, e.g., Cross, supra, 14 Cal.App.5th at pp. 206–207 [plaintiffs’ breach of contract claim barred by CDA]; Federal Agency of News, supra, 432 F.Supp.3d at pp. 1119–1120 [plaintiffs’ causes of action for breach of contract and breach of the implied covenant of good faith and fair dealing prohibited by § 230]; Brittain v. Twitter, Inc., supra, 2019 WL 2423375 at pp. *3–*4 [dismissing, among others, causes of action for breach of contract and promissory estoppel]; King v. Facebook, Inc. (N.D.Cal. Sept. 5, 2019, No. 19-cv-01987-WHO) 2019 WL 4221768 at pp. *1, fn. 1, *3–*5 [dismissing breach of contract, promissory estoppel, and UCL claims under § 230(c)(1)]; Caraccioli v. Facebook, Inc. (N.D.Cal. 2016) 167 F.Supp.3d 1056, 1061, 1064–1066 [dismissing breach of contract and UCL claims under

17 § 230]; Goddard v. Google, Inc. (N.D.Cal. 2009) 640 F.Supp.2d 1193, 1199, 1201 (Goddard) [dismissing breach of contract claim based on search engine’s failure to abide by terms of its content policy].) Murphy relies heavily on Barnes, supra, 570 F.3d 1096 to argue the viability of her claims, but that case is distinguishable. In Barnes, the Ninth Circuit considered whether an Internet service provider was liable for its failure to remove material harmful to the plaintiff but failed to do so. Plaintiff Barnes had broken off a long relationship with her boyfriend, who then posted unauthorized false profiles of her to Yahoo, Inc.’s (Yahoo) website, soliciting sex. She repeatedly demanded that Yahoo remove the false profiles, but the company did not respond. The day before a local news story about the incident was to be broadcast, Yahoo’s director of communications called Barnes and told her she would “ ‘personally walk the statements over to the division responsible for stopping unauthorized profiles and they would take care of it.’ ” (Id. at pp. 1098–1099.) Barnes relied on Yahoo’s promise to remove the content, but Yahoo still did not take down the profiles until after Barnes filed her lawsuit. (Id. at p. 1099.) Barnes filed a complaint against Yahoo for negligent undertaking and promissory estoppel. (Barnes, supra, 570 F.3d at p. 1099.) The court concluded Barnes’s negligent undertaking claim was barred by section 230(c)(1) because “removing content is something publishers do,” and the “duty that Barnes claims Yahoo violated derives from Yahoo’s conduct as a publisher—the steps it allegedly took, but later supposedly abandoned, to de-publish the offensive profiles.” (Barnes, at p. 1103.) By contrast, the court held, her promissory estoppel claim was not precluded because “the duty the defendant allegedly violated springs from a contract—an enforceable promise—not from any non-contractual conduct or capacity of the defendant.

18 [Citation.] Barnes does not seek to hold Yahoo liable as a publisher or speaker of third-party content, but rather as the counter-party to a contract, as a promisor who has breached.” (Id. at p.1107.) Barnes never suggested, however, that all contract or promissory estoppel claims survive CDA immunity. (See, e.g., Cross, supra, 14 Cal.App.5th at p. 207 [noting Barnes directs us to look beyond the name of the cause of action and examine whether the duty plaintiff alleges defendant violated derives from defendant’s status as a publisher to determine whether § 230(c)(1) precludes liability]; Goddard, supra, 640 F.Supp.2d at p. 1200 [“Read as broadly as possible, Barnes stands for the proposition that when a party engages in conduct giving rise to an independent and enforceable contractual obligation, that party may be ‘h[eld] . . . liable . . . as a counter- party to a contract, as a promisor who has breached.’ ”].) Unlike in Barnes, where the plaintiff sought damages for breach of a specific personal promise made by an employee to ensure specific content was removed from Yahoo’s website, the substance of Murphy’s complaint accuses Twitter of unfairly applying its general rules regarding what content it will publish and seeks injunctive relief to demand that Twitter restore her account and refrain from enforcing its Hateful Conduct Policy. Murphy does not allege someone at Twitter specifically promised her they would not remove her tweets or would not suspend her account. Rather, Twitter’s alleged actions in refusing to publish and banning Murphy’s tweets, as the trial court in this case observed, “reflect paradigmatic editorial decisions not to publish particular content” that are protected by section 230. Indeed, the Barnes court itself recognized a difference between the type of allegations Murphy makes here and the specific promise alleged in that case. Noting that “as a matter of contract law, the promise must ‘be as clear

19 and well defined as a promise that could serve as an offer, or that otherwise might be sufficient to give rise to a traditional contract supported by consideration,” the Barnes court explained that “a general monitoring policy . . . does not suffice for contract liability.” (Barnes, supra, 570 F.3d at p. 1108.) Here, Murphy’s allegations that Twitter “enforced its Hateful Conduct Policy in a discriminatory and targeted manner” against Murphy and others by removing her tweets and suspending her account amount to attacks on Twitter’s interpretation and enforcement of its own general policies rather than breach of a specific promise.6 We likewise are not persuaded by Murphy’s reliance on Demetriades, supra, 228 Cal.App.4th 294. There, a restaurant operator sought an injunction under the UCL to prevent Yelp, Inc. (Yelp) from making claims

6 Although Murphy also points to the allegations that Twitter failed to give her 30 days’ notice of the changes to the Hateful Conduct Policy and that Twitter applied its new policy retroactively as breaches of clear and well- defined promises, the gravamen of each of her causes of action concerns Twitter’s editorial decisions not to publish content—as reflected by the fact that she alleges no specific injury from the alleged notice and retroactivity violations but complains instead of the harm caused by Twitter’s ban on her and others’ free speech rights. (See, e.g., Noah v. AOL Time Warner, Inc. (E.D.Va. 2004) 261 F.Supp.2d 532, 538, affd. 2004 WL 602711 (4th Cir. 2004) [CDA barred discrimination claim because examination of injury plaintiff claimed and remedy he sought clearly indicated his claim sought to place defendant in publisher’s role].) Moreover, Twitter’s Hateful Conduct Policy as adopted in December 2017, before Murphy posted any of her allegedly offending tweets, broadly disallowed “behavior that harasses individuals,” and expressly proscribed “directly attack[ing] . . . other people on the basis of . . . gender identity” and “repeated and/or non-consensual slurs, epithets, racist and sexist tropes, or other content that degrades someone.” We fail to see how Twitter’s amendment of its policy in October 2018 to include “targeted misgendering or deadnaming” as illustrative examples of such behavior constitute a change altering any party’s “rights or obligations” sufficient to give rise to a breach of contract claim based on Twitter’s user agreement.

20 about the accuracy of its program that filters restaurant reviews. (Demetriades, at pp. 299–300.) Yelp filed a motion to strike the complaint under Code of Civil Procedure section 425.16 (anti-SLAPP motion) and the court concluded the commercial speech exception under Code of Civil Procedure section 425.17 applied to Yelp’s statements about the accuracy and efficacy of its filter. (Demetriades, at pp. 305, 310.) The opinion contains very little discussion of the CDA, but the court determined Yelp was not immune under section 230 because the plaintiff sought to hold Yelp liable for its own “specific and detailed” factual statements about the accuracy of its filter. (Demetriades, at pp. 311, 313.) Here, the gravamen of Murphy’s complaint seeks to hold Twitter liable, not for specific factual representations it made, but for enforcing its Hateful Conduct Policy against her and exercising its editorial discretion to remove content she had posted on its platform.7 Nor are we persuaded by Murphy’s argument that section 230 immunity does not apply here because the content at issue is Twitter’s own promises rather than content generated by Murphy and other users. Courts have repeatedly determined that when plaintiffs allege a platform has wrongfully ceased publishing their posts or blocked content, that content constitutes “information provided by another information content provider”

7 Murphy also cites Joude v. WordPress Foundation (N.D.Cal. Jul. 3, 2014, No. C14-01656 LB) 2014 WL 3107441, to support her contract claim, but that case is particularly unpersuasive. Murphy quotes the court’s statement that “claim two is a contract claim that—if viable—would be a state law claim that would not be barred by section 230,” but the Joude court had analyzed in great detail how the plaintiffs failed to allege a cognizable cause of action for breach of contract based on the provider’s terms of service, noted it could not “conceive of an amendment that would cure the shortcomings,” and expressly stated the “claim looks to be a claim that ought to be made against the blogger, however, not [the interactive service provider].” (Joude, at pp. *8, *6.)

21 within the meaning of section 230(c)(1). (See Sikhs for Justice “SFJ”, Inc. v. Facebook, Inc., supra, 144 F.Supp.3d at pp. 1090, 1093–1094 [explaining that § 230(c)(1)’s reference to “another information content provider” refers to an interactive computer service provider that passively displays third party content as opposed to one that participates in creating or developing it]; Federal Agency of News, supra, 432 F.Supp.3d at pp. 1117–1118 [defendant satisfied “another information content provider” prong of § 230 immunity test where plaintiffs sought to hold defendant liable for removing plaintiffs’ account, posts, and content].) Division Two of this court rejected a claim similar to Murphy’s in Cross, supra, 14 Cal.App.5th at pages 206–207. There, the plaintiff argued that portions of statements made in “Facebook’s terms and community standards” were “ ‘representations of fact’ made by Facebook” and that section 230 did not apply to the breach of contract and negligence claims because the complaint “ ‘specifically allege[d] that Facebook is liable because of its own promises and representations to [plaintiff], not because of anyone else’s statements.’ ” (Id. at pp. 203, 206.) The Cross court rejected the plaintiff’s argument that Facebook was liable for failing “ ‘to adhere to its own legally enforceable promise.’ ” (Cross, supra, 14 Cal.App.5th at p. 207.) Citing Barnes, it explained that “[i]n evaluating whether a claim treats a provider as a publisher or speaker of user-generated content, ‘what matters is not the name of the cause of action,’ ” but “ ‘whether the cause of action inherently requires the court to treat the defendant as the “publisher or speaker” of content provided by another.’ ” (Cross, at p. 207.) The court then cited numerous state and federal court decisions that have found claims based on a failure to remove content posted by others were barred by the CDA. (Cross, at p. 207.)

22 We agree with the Cross court’s analysis, and find it equally applicable in this case. Murphy argues Cross is distinguishable because it involved liability for a service provider’s failure to remove third party content, while she seeks to hold Twitter liable for its own promises and representations. But, as we have discussed, that is the very same argument the Cross court considered and rejected, i.e., that Twitter can be held liable based on promises and representations in its general terms of service. (Cross, supra, 14 Cal.App.5th at p. 207.) Murphy also notes the plaintiff in Cross failed to “ ‘identif[y] any “representation of fact” that Facebook would remove any objectionable content,’ ” whereas she identified several terms in the user agreement. But Murphy does not identify any specific representation of fact or promise by Twitter to Murphy that it would not remove her tweets or suspend her account beyond general statements in its monitoring policy, the type of allegation the Barnes court noted would be insufficient to state a claim. (Barnes, supra, 570 F.3d at p. 1108.) Murphy also urges us to conclude that the trial court erred because section 230(c)(2), not section 230(c)(1), governs Twitter’s actions suspending her account and removing her tweets from its platform.8 Section 230(c)(2) provides, in relevant part: “No provider or user of an interactive computer service shall be held liable on account of— [¶] (A) any action voluntarily taken in good faith to restrict access to or availability of material that the provider or user considers to be obscene, lewd, lascivious, filthy, excessively violent, harassing, or otherwise objectionable, whether or not such material is constitutionally protected.” Murphy contends the clear text of the statute,

8 Twitter expressly confirmed at the trial court hearing that it was not relying on section 230(c)(2) for purposes of the demurrer to Murphy’s complaint.

23 the canon of statutory interpretation against surplusage, and our Supreme Court’s decision in Barrett, supra, 40 Cal.4th 33, all preclude Twitter from relying on section 230(c)(1), as opposed to section 230(c)(2), to defend its removal of allegedly harassing content. We are not convinced. As the Ninth Circuit explained in Barnes, section 230(c)(1) “shields from liability all publication decisions, whether to edit, to remove, or to post, with respect to content generated entirely by third parties.” (Barnes, supra, 570 F.3d at p. 1105, italics added.) Section 230(c)(2), on the other hand, applies “not merely to those whom subsection (c)(1) already protects, but any provider of an interactive computer service” regardless of whether the content at issue was created or developed by third parties. (Barnes, at p. 1105.) Thus, section 230(c)(2) “provides an additional shield from liability,” encompassing, for example, those interactive computer service providers “who cannot take advantage of subsection (c)(1) . . . . because they developed, even in part, the content at issue.” (Barnes, at p. 1105, italics added.) The Ninth Circuit recently reiterated that analysis in Fyk v. Facebook, supra, 808 Fed.Appx. 597 (Fyk), holding that section 230(c)(1) immunized the interactive computer service based on its alleged “de-publishing” and “re- publishing” of user content. (Fyk, at pp. 597–598.) The Ninth Circuit explicitly rejected the argument that applying section 230(c)(1) immunity to such decisions renders section 230(c)(2) “mere surplusage.” (Fyk, at p. 598.) Other federal courts are in agreement.9 (See Domen v. Vimeo, Inc., supra,

9 Murphy cites Zango, Inc. v. Kaspersky Lab, Inc. (9th Cir. 2009) 568 F.3d 1169, 1175 and e-ventures Worldwide, LLC v. Google, Inc. (M.D.Fla. Feb. 8, 2017, No. 2:14-cv-646-FtM-PAM-CM) 2017 WL 2210029 in support of her argument that section 230(c)(1) and (2) “apply to different concerns.” Zango, however, is inapposite because it did not address whether

24 433 F.Supp.3d at p. 603 [applying § 230(c)(1) to a platform’s decision to refuse to publish content does not render § 230(c)(2)’s good faith requirement surplusage because “[s]ection 230(c)(2)’s grant of immunity, while ‘overlapping’ with that of [s]ection 230(c)(1), [citation], also applies to situations not covered by [s]ection 230(c)(1)”]; Force v. Facebook, Inc. (2d Cir. 2019) 934 F.3d 53, 79–80 (conc. opn. of Katzmann, C. J.) [noting § 230(c)(2) provides for much narrower civil liability than broad liability under § 230(c)(1), which covers publishers’ decisions regarding content removal].) Contrary to Murphy’s argument, this interpretation of the statutory scheme is not inconsistent with our Supreme Court’s analysis in Barrett. Barrett did not involve content restrictions by an interactive computer service provider, and thus the Supreme Court had no occasion to hold that only section 230(c)(2) applies to such claims. Rather, the court considered whether section 230 confers immunity not only on “ ‘publishers,’ ” but also common law “ ‘distributors’ ” of allegedly defamatory material. (Barrett, supra, 40 Cal.4th at p. 39.) In its discussion of section 230(c)(1) and (2), the

section 230(c)(1) immunity applies to decisions to remove or block content, but whether a company that provided malware-blocking software to users was protected from liability under section 230(c)(2)(B) for blocking access to the plaintiff’s programs. (Zango, at pp. 1174–1175 [“this case presents a different problem, and a statutory provision with a different aim, from ones we have encountered before”].) While e-Ventures supports Murphy’s argument here, we find it unpersuasive because the court did not consider the distinction recognized by courts in the Ninth Circuit between a publisher that is involved in the creation of content and one who only makes publication decisions regarding content created by others. We agree with the Ninth Circuit’s analysis in Barnes and Fyk that reading section 230(c)(1) to cover all publication decisions with respect to content generated entirely by third parties does not render section 230(c)(2) mere surplusage, because that section provides an additional shield from liability when interactive computer services play some role in the creation or development of content. (Barnes, supra, 570 F.3d at p. 1105; Fyk, supra, 808 Fed.Appx. at p. 598.)

25 Supreme Court disagreed with the Court of Appeal’s reasoning that section 230(c)(2) would be superfluous if “ ‘publishers’ ” and “ ‘distributors’ ” of third party content both had broad immunity under section 230(c)(1). (Barrett, at pp. 48–49.) Noting that the “terms of section 230(c)(1) are broad and direct,” the court rejected an interpretation of the CDA that assumed Congress intended to immunize “ ‘publishers’ ” but leave “ ‘distributors’ ” vulnerable to liability.10 (Barrett, at pp. 47–48.) Importantly, Barrett also expressly agreed with the Zeran court’s analysis of section 230 immunity and its construction of the term “publisher.” (Barrett, supra, 40 Cal.4th at pp. 41–42, 48–49.) As discussed above, Murphy’s claims here fall squarely within Zeran’s holding that “lawsuits seeking to hold a service provider liable for its exercise of a publisher’s traditional editorial functions—such as deciding whether to publish, withdraw, postpone or alter content—are barred.” (Zeran, supra, 129 F.3d at p. 330.) Moreover, the Barrett court acknowledged in its discussion of the legislative history of section 230 that the statute was “enacted to remove the disincentives to self-regulation created by [Stratton Oakmont, supra,

10 We acknowledge some ambiguity in our high court’s statements that section 230(c)(1) and (2) “address different concerns” and that section 230(c)(1) “is concerned with liability arising from information provided online,” while section 230(c)(2) is “directed at actions taken by Internet service providers or users to restrict access to online information.” (Barrett, supra, 40 Cal.4th at p. 49.) Thus, “[s]ection 230(c)(1) provides immunity from claims by those offended by an online publication, while section 230(c)(2) protects against claims by those who might object to the restriction of access to an online publication.” (Ibid.) We disagree with Murphy, however, that the statements suggest only section 230(c)(2) applies to her claims. Rather, we agree with Twitter and amicus curiae that the passage is dicta, and read in context of our high court’s broader analysis and specific holdings, conclude it cannot save Murphy’s claims here.

26 23 Media L.Rep. 1794 [1995 WL 323710] ], in which a service provider was held liable as a primary publisher because it actively screened and edited messages posted on its bulletin boards.” (Barrett, supra, 40 Cal.4th at p. 51.) The court recognized that comments made by Representative Cox, one of the two sponsors of section 230, meant that distributors would be “protected from rather than threatened with liability, to encourage responsible screening of the content provided on their services.” (Barrett, at p. 53.) It further explained that both the “terms of section 230(c)(1) and the comments of Representative Cox reflect the intent to promote active screening by service providers of online content provided by others.” (Barrett, at p. 53, italics added.) “Congress contemplated self-regulation, rather than regulation compelled at the sword point of tort liability.” (Ibid.) Finally, we reject Murphy’s arguments that the superior court’s ruling frustrates section 230’s stated policy purposes. As our Supreme Court explained in Barrett, one of the key purposes of section 230 “was ‘to encourage service providers to self-regulate the dissemination of offensive material over their services.’ [Citation.] . . . ‘Fearing that the specter of liability would . . . deter service providers from blocking and screening offensive material, Congress enacted § 230’s broad immunity,’ which ‘forbids the imposition of publisher liability on a service provider for the exercise of its editorial and self-regulatory functions.’ ” (Barrett, supra, 40 Cal.4th at p. 44, fn. omitted.) Based on the clear statutory objectives as defined by Congress in the statute itself and our own Supreme Court’s guidance, we conclude the superior court’s ruling does not interfere with section 230’s twin policy goals of promoting free and open online discourse while encouraging Internet computer services to engage in active monitoring of offensive content. (Hassell, supra, 5 Cal.5th at p. 534.)

27 In sum, the trial court properly sustained Twitter’s demurrer without leave to amend because each of Murphy’s claims is barred under section 230(c)(1). B. Failure to State a Claim Even assuming, however, that section 230(c)(1) immunity does not apply, we would affirm the trial court’s judgment because Murphy failed to state a cognizable cause of action. 1. Breach of Contract Murphy’s cause of action for breach of contract is based on Twitter’s failure to provide notice of changes to its Hateful Conduct Policy, its retroactive enforcement of the policy, and its breach of the covenant of good faith and fair dealing because it “targeted her for permanent suspension despite the fact that she never violated any of the Terms of Service, Rules or incorporated policies.” Her claim necessarily fails, however, because Twitter’s terms of service expressly state that they reserve the right to “suspend or terminate [users’] accounts . . . for any or no reason” without liability. (See Cox v. Twitter (D.S.C. 2019) 2019 WL 2513963 at p. *4 [plaintiff’s breach of contract claim against Twitter for violation of its terms of service in removing content and suspending his account were clearly barred by terms of user agreement]; Ebeid v. Facebook, Inc., supra, 2019 WL 2059662 at p. *8 [breach of implied covenant of good faith and fair dealing failed because Facebook had contractual right to remove any post at Facebook’s sole discretion].) Waivers of liability that are “ ‘clear, unambiguous and explicit’ ” bar claims that expressly fall within their scope. (Cohen v. Five Brooks Stable (2008) 159 Cal.App.4th 1476, 1485.) The clear terms of Twitter’s user agreement

28 preclude a claim for breach of contract based on the allegations of Murphy’s complaint. Murphy does not dispute that her claims fall within the scope of Twitter’s liability waiver provision but contends that the terms are unconscionable and that the court should strike those provisions and enforce the remainder of the contract. Murphy fails, however, to allege any facts demonstrating the terms are procedurally or substantively unconscionable. A contract term may be substantively unconscionable if it is “ ‘ “ ‘ “overly harsh” ’ ” [citation], “ ‘unduly oppressive’ ” [citation], “ ‘so one- sided as to “shock the conscience” ’ ” [citation], or “unfairly one-sided.” ’ ” (Baltazar v. Forever 21, Inc. (2016) 62 Cal.4th 1237, 1244.) Murphy alleges Twitter’s terms of service are substantively unconscionable because Twitter could use them to suspend or terminate user accounts for petty, arbitrary, irrational, discriminatory, or unlawful reasons. Terms allowing service providers to “discontinue service, or remove content unilaterally,” however, are routinely found in standardized agreements and enforced by courts. (Song fi, Inc. v. Google, Inc. (D.D.C. 2014) 72 F.Supp.3d 53, 63 (Song fi, Inc.).) “Unless there is some evidence of ‘egregious’ tactics, . . . ‘the party seeking to avoid the contract will have to show that the terms are so extreme as to appear unconscionable according to the mores and business practices of the time and place.’ ” (Ibid.) Courts have also recognized service providers that offer free services to Internet users may have a legitimate commercial need to limit their liability and have rejected claims that such limitations are so one-sided as to be substantively unconscionable. (See Lewis v. YouTube, LLC (2015) 244 Cal.App.4th 118, 125–126 [limitation of liability clauses “are appropriate when one party is offering a service for free to the public”; plaintiff failed to

29 state a claim for breach of contract because YouTube, LLC’s (YouTube) terms of service limited its liability for deleting her video]; Darnaa, LLC v. Google, LLC (9th Cir. 2018) 756 Fed.Appx. 674, 676 [YouTube’s terms of service not substantively unconscionable because it offers video streaming services at no cost to user]; see also Song fi, Inc., supra, 72 F.Supp.3d at pp. 63–64 [finding YouTube had legitimate commercial need to include forum selection clause].) In light of Murphy’s allegations that Twitter provides its services to millions of users around the world for free, the contract term allowing it to suspend or terminate users’ accounts for any or no reason, absent other factual allegations of unfairness, does not shock the conscience or appear unfairly one-sided. Murphy makes other conclusory allegations of substantive unconscionability based on quotes from an arbitration case, Sanchez v. Valencia Holding Co., LLC (2015) 61 Cal.4th 899, 911, asserting Twitter’s terms of service are “ ‘unreasonably favorable to the more powerful party’ ” and “ ‘unfairly one-sided.’ ” She further alleges, “The terms purporting to give Twitter the right to suspend or ban an account ‘at any time for any or no reason’ and ‘without liability to you’ ‘contravene the public interest or public policy,’ ‘attempt to alter in an impermissible manner fundamental duties otherwise imposed by the law,’ ‘seek to negate the reasonable expectations of the nondrafting party,’ and impose ‘unreasonably and unexpectedly harsh terms having nothing to do with . . . central aspects of the transaction.’ ” But Murphy offers no specific factual allegations in support of these contentions. In her reply brief on appeal, Murphy relies on In re Yahoo! Customer Data Sec Breach Litigation (N.D.Cal. 2018) 313 F.Supp.3d 1113, 1138, in support of her unconscionability argument, but that case does not assist her. In In re Yahoo!, the court concluded the plaintiffs had sufficiently pleaded a

30 claim for substantive unconscionability “[u]nder the particular circumstances of this case,” where they alleged the defendants had legal obligations under both state and federal law to maintain acceptable levels of data security, the plaintiffs suffered actual damages, and the defendants were better equipped to maintain secure systems than individual email users, resulting in an allocation of risk that was unreasonable and unexpected. (Id. at pp. 1137– 1138.) Murphy made no similar allegations of legal obligations on the part of Twitter, or of damages suffered by Murphy and other users, with respect to Twitter’s waiver of liability for removing content and suspending accounts. Murphy also relies on a Pennsylvania case applying California law, Bragg v. Linden Research, Inc. (E.D.Pa. 2007) 487 F.Supp.2d 593, 595–596, which is also distinguishable. In Bragg, a Pennsylvania federal court applying California law concluded that an arbitration agreement was substantively unconscionable because the terms of service provided the defendant “with a variety of one-sided remedies to resolve disputes, while forcing its customers to arbitrate any disputes with [it].” (Id. at p. 608.) Specifically, the company seized the plaintiff’s account, retained funds he had paid them, then told him he could resolve the dispute by initiating a costly arbitration process. (Ibid.) The court also concluded the defendant’s retention of the unilateral right to modify the arbitration agreement evidenced a lack of mutuality supporting a finding of substantive unconscionability, found other terms of the arbitration agreement unfair, and concluded the contract was procedurally unconscionable based on both its adhesive nature and surprise. (Id. at pp. 606–611.) No similar facts are alleged here. Murphy has also failed to allege facts demonstrating the terms were procedurally unconscionable. While the contract is an adhesion contract,

31 such contracts are “ ‘indispensable facts of modern life that are generally enforced.’ ” (Baltazar v. Forever 21, Inc., supra, 62 Cal.4th at p. 1244; see AT&T Mobility LLC v. Concepcion (2011) 563 U.S. 333, 346–347 [“the times in which consumer contracts were anything other than adhesive are long past”].) The fact that Murphy had no opportunity to negotiate the terms of service, standing alone, is insufficient to plead a viable unconscionability claim. (Gatton v. T-Mobile USA, Inc. (2007) 152 Cal.App.4th 571, 585 [“use of a contract of adhesion establishes a minimal degree of procedural unconscionability notwithstanding the availability of market alternatives. If the challenged provision does not have a high degree of substantive unconscionability, it should be enforced.”]; see Sweet v. Google, Inc. (N.D.Cal. Mar. 7, 2018, No. 17-cv-03953-EMC) 2018 WL 1184777 at p. *5 [“Even if the contract(s) at issue were deemed adhesive, that simply establishes a minimal degree of procedural unconscionability such that [the plaintiff] would have to establish a fair amount of substantive unconscionability in order to prevail.”].) Murphy does not allege any facts demonstrating surprise, such as “the ‘terms of the bargain [being] hidden in a prolix printed form’ or pressure to hurry and sign.” (De La Torre v. CashCall, Inc. (2018) 5 Cal.5th 966, 983.) Accordingly, we conclude she had pleaded facts showing only a minimal degree of procedural unconscionability. Murphy’s failure to plead any other facts supporting her unconscionability analysis and her failure to demonstrate an ability to amend to plead a cognizable cause of action, leads us to affirm the trial court on this additional ground. 2. Promissory Estoppel A claim for promissory estoppel requires (1) a promise clear and unambiguous in its terms, (2) reliance by the party to whom the promise is made, (3) the reliance must be reasonable and foreseeable, and (4) the party

32 asserting the estoppel must be injured by his or her reliance. (Flintco Pacific, Inc. v. TEC Management Consultants, Inc. (2016) 1 Cal.App.5th 727, 734.) Murphy alleges she relied on the following “clear and unambiguous” promises by Twitter: (1) the “Twitter Rules” applicable when Murphy joined stated, “ ‘we do not actively monitor user’s content and will not censor user content,’ except in limited circumstances such as impersonation, violation of trademark or copyright, or ‘direct specific threats of violence against others’ ”; (2) Twitter’s terms of service promised Murphy she would receive 30 days’ notice of any changes to its policies; (3) Twitter’s terms of service stated it would not enforce its policies retroactively against her; (4) Twitter’s “Enforcement Guidelines” stated that “ ‘account-level’ actions” are reserved “for cases where ‘a person has violated the Twitter Rules’ ” repeatedly or in a particularly egregious way; (5) Twitter’s “Safety page” states, “ ‘We treat everyone equally: the same Twitter Rules apply to all’ and ‘You have the right to express yourself on Twitter if you adhere to these rules’ ”; and (6) Twitter’s CEO, Jack Dorsey, testified before Congress that Twitter does not “ ‘consider political viewpoints, perspectives, or party affiliation in any of our policies or enforcement decisions, period.’ ” Murphy alleges she reasonably and foreseeably relied on these promises to her detriment, and has lost valuable economic interests in access to her Twitter account and followers forever. None of these alleged promises, however, suffice to state a claim because Murphy could not reasonably rely on promises that Twitter would not restrict access to her account, “ ‘censor’ ” her content, or take “account- level” action when the terms of service stated at all relevant times that Twitter could “ ‘remove or refuse to distribute any Content’ ” and could suspend or terminate an account “ ‘for any or no reason.’ ” “ ‘ “[W]hether a

33 party’s reliance was justified may be decided as a matter of law if reasonable minds can come to only one conclusion based on the facts.” ’ ” (Daniels v. Select Portfolio Servicing, Inc. (2016) 246 Cal.App.4th 1150, 1179; Joffe v. City of Huntington Park (2011) 201 Cal.App.4th 492, 513.) No plausible reading of the six contract terms and general statements Murphy identifies promises users that they will not have their content removed nor lose access to their account based on what they post online. Indeed, the very terms of service that Murphy relies on in asserting her claims make clear that Twitter may suspend or terminate an account for any or no reason. (See Malmstrom v. Kaiser Aluminum & Chemical Corp. (1986) 187 Cal.App.3d 299, 318–319 [rejecting promissory estoppel claim because reliance on representations contradicting written agreement is not reasonable].) Moreover, the operative Hateful Conduct Policy in place in December 2017, when Murphy began posting her deleted tweets, stated that Twitter “do[es] not tolerate behavior that harasses, intimidates, or uses fear to silence another person’s voice,” and offered examples of intolerable behavior “includ[ing], but not limited to,” “repeated and/or non-consensual slurs, . . . sexist tropes, or other content that degrades someone.” The breadth and ambiguity of this prohibition makes any reliance on vague statements that Twitter will not “censor” content unreasonable. Because Murphy has not alleged Twitter ever made a specific representation directly to her or others that they would not remove content from their platform or deny access to their accounts, but rather expressly reserved the right to remove content, including content they determine is harassing or intolerable, and suspend or terminate accounts “for any or no reason” in its terms of service, Murphy cannot plead reasonable reliance on the alleged promises as a matter of law.

34 3. Violation of the UCL Murphy’s claim for violation of the UCL alleges that Twitter engaged in unfair and fraudulent business practices. As an initial matter, Murphy’s UCL claim fails for lack of standing because she does not allege a “loss or deprivation of money or property sufficient to qualify as injury in fact, i.e., economic injury.” (Kwikset Corp. v. Superior Court (2011) 51 Cal.4th 310, 322.) Murphy’s complaint alleges that she and others “lost a tangible property interest in their accounts and the followers they had accumulated,” that she is a “freelance journalist and writer who relies on Twitter for her livelihood,” and that “[t]here is no public forum comparable to Twitter that would allow Murphy and other users to build a widespread following, communicate with a global audience, or support themselves in the fields of journalism, politics, or public affairs.” Murphy also alleges that without her Twitter account she is “unable to . . . share links to her Patreon account (where readers can support her work financially).” Murphy does not dispute that under Twitter’s terms of service, however, she does not have a property interest in her account or her followers, but only the content she creates. And while she alleges generally that she relies on Twitter for her livelihood, she does not allege she suffered any actual loss of income or financial support. Murphy’s theory that Twitter engaged in “unfair” behavior is premised on the affirmative claim that inclusion of its liability waiver provisions in its terms of service was unconscionable. For the reasons discussed above, we reject that theory of liability based on the allegations of Murphy’s complaint.11

Because we conclude Murphy has not stated a claim for 11

unconscionability, we need not consider Twitter’s argument that the UCL

35 Murphy also asserts that Twitter engaged in fraudulent activity “because it held itself out to be a free speech platform” on its website, and in advertising, public statements, and its CEO’s testimony before Congress. Murphy relies on the same statements and contractual provisions asserted in her breach of contract and promissory fraud claims. Murphy claims she and others “reasonably assumed that Twitter would allow them to use the forums to freely express their opinions on all subjects, without engaging in censorship based on their political views and affiliations, so long as they did not threaten or harass others.” Murphy’s allegations that Twitter’s general declarations of commitment to free speech principles cannot support a fraud claim, because it is unlikely that members of the public would be deceived by such statements. (Kwikset Corp. v. Superior Court, supra, 51 Cal.4th at p. 326 [fraud prong requires actual reliance on the allegedly deceptive or misleading statements]; Shaeffer v. Califia Farms, LLC (2020) 44 Cal.App.5th 1125, 1140 [whether consumers are likely to be deceived may be resolved on demurrer if facts alleged fail to show as matter of law that a reasonable consumer would be misled].) No reasonable person could rely on proclamations that “[w]e believe in free expression and think every voice has the power to impact the world,” that Twitter was the “free speech wing of the free speech party,” or that Twitter’s mission “ ‘is to give everyone the power to create and share ideas and information instantly without barriers,’ ” as a

does not permit such claims. (Compare Rubio v. Capital One Bank (9th Cir. 2010) 613 F.3d 1195, 1205 [“Under California law, . . . unconscionability is an affirmative defense [citation] not a cause of action”] with California Grocers Assn. v. Bank of America (1994) 22 Cal.App.4th 205, 218 [court “assumed” UCL “encompass[ed] an affirmative cause of action for unconscionability”]; De La Torre v. CashCall, Inc., supra, 5 Cal.5th at p. 980 [citing cases assuming UCL encompasses affirmative cause of action for unconscionability].)

36 promise that Twitter would not take any action to self-regulate content on its platform. (See, e.g., Prager University v. Google LLC (9th Cir. 2020) 951 F.3d 991, 1000 [“YouTube’s braggadocio about its commitment to free speech constitutes opinions” and “[l]ofty but vague statements” about allowing individuals to speak freely “are classic, non-actionable opinions or puffery”].) As to the provisions in Twitter’s user agreement on which Murphy bases her breach of contract and promissory estoppel claims, they, too, are not likely to mislead members of the public as a matter of law. No reasonable person would rely on a general statement that Twitter would not “actively monitor user’s content and will not censor user content, except in limited circumstances” to mean it would never restrict content, particularly when the same rules offered examples of impermissible conduct, reserved the right to change the rules, and gave Twitter “the right to immediately terminate [an] account without further notice in the event that, in its judgment, [a user] violate[s] [the Twitter Rules] or the Terms of Service.” Because Murphy has failed to allege representations by Twitter that were likely to deceive or mislead members of the public, her UCL claim fails. 4. Leave to Amend Finally, Murphy contends the trial court erred in sustaining the demurrer without leave to amend. She argues the trial court “suggested that Murphy’s claims would not be barred by [section] 230(c)(1) if she merely sought ‘damages for Twitter’s failure to comply with an alleged contractual or quasi-contractual promise.’ ” Murphy argues she should have been granted leave to amend to “substitute the relief that the Superior Court deemed consistent with [section] 230(c)(1).” Contrary to Murphy’s argument, the trial court did not suggest Murphy’s claims would be viable; rather, it concluded they were barred by

37 section 230 immunity. In a footnote, the trial court wrote that Murphy’s case was distinguishable from the facts of Barnes because she “is not seeking damages for Twitter’s failure to comply with an alleged contractual or quasi- contractual promise, but rather is seeking injunctive relief to compel it to restore her and others’ Twitter accounts and to refrain from enforcing its Hateful Conduct Policy against her.” Despite Murphy’s conclusory statement that she should be given leave to amend, she offered no explanation in the trial court or in her briefing on appeal of what specific promise or contractual provision Twitter violated that would result in a claim for damages or what those damages would be. Murphy correctly advances the principle that leave to amend may be requested for the first time on appeal. (Code Civ. Proc., § 472c, subd. (a).) She fails, however, to explain how she could amend to allege a cognizable claim that would survive section 230(c)(1)’s broad immunity or remedy the defects in her causes of action discussed above. “ ‘[T]he burden of showing that a reasonable possibility exists that amendment can cure the defects remains with the plaintiff; neither the trial court nor this court will rewrite a complaint. [Citation.] Where the appellant offers no allegations to support the possibility of amendment and no legal authority showing the viability of new causes of action, there is no basis for finding the trial court abused its discretion when it sustained the demurrer without leave to amend.’ ” (Total Call Internat., Inc. v. Peerless Ins. Co. (2010) 181 Cal.App.4th 161, 173.) Accordingly, we affirm the judgment of dismissal. C. First Amendment Because we hold each of Murphy’s claims is barred by section 230(c)(1) and Murphy has failed to state a cause of action under California law, we find it unnecessary to address Twitter’s argument that Murphy’s claims violate

38 the First Amendment. (See Hassell, supra, 5 Cal.5th at p. 534 [courts avoid resolving constitutional questions if the issue may be resolved on narrower grounds].) D. Request for Judicial Notice Murphy requested we take judicial notice of (1) a legal ruling from a Canadian judicial tribunal; (2) the prepared remarks of Federal Trade Commissioner Rohit Chopra for the spring meeting of the American Bar Association in 2019; and (3) a settlement agreement in Dept. of Fair Empl. & Housing v. AirBnB, Inc. (2017) Nos. 574743-231889, 574743-231624, Voluntary Agreement (as of January 22, 2021). The legal ruling from a Canadian tribunal rejecting discrimination complaints filed by Jessica Y. is irrelevant to the resolution of this appeal. Whether those claims had merit is immaterial to determining whether Twitter is entitled to immunity under section 230, or whether Murphy has stated a claim for relief. We take judicial notice, however, that the opinion refers to one of the individuals Murphy tweeted about, Jessica Y., as a “transgender woman.” We deny the request for judicial notice of the Federal Trade Commissioner’s remarks to the American Bar Association that an overly broad reading of section 230 could undermine antitrust enforcement efforts because it has no relevance to this appeal, which concerns breach of contract and promissory estoppel claims. We deny Murphy’s request that we take judicial notice of the California Department of Fair Employment and Housing’s settlement with AirBnB, Inc. outlining policy concerns regarding interactive computer service providers

39 claiming immunity from antidiscrimination laws, which again, are not at issue in this appeal.

III. DISPOSITION

The judgment is affirmed. Defendants are to recover their costs on appeal.

40

MARGULIES, J.

WE CONCUR:

HUMES, P. J.

BANKE, J.

A158214

Murphy v. Twitter, Inc.

41 Trial Court: San Francisco Superior Court

Trial Judge: Ethan P. Schulman

Counsel:

Dhillon Law Group, Inc., Harmeet K. Dhillon; Michael Yamamoto LLC and Gregory R. Michael for Plaintiff and Appellant.

Wilmer Cutler Pickering Hale and Dorr LLP, Patrick J. Carome, Ari Holtzblatt and Thomas G. Sprankling for Defendants and Respondents.

Jenner & Block LLP, Luke C. Platzer; National Center for Lesbian Rights, Shannon Minter, Asaf Orr; Zeb C. Zankel; Ethan C. Wong; Vaughn E. Olson; GLBTQ Advocates & Defenders and Jennifer Levi for National Center for Lesbian Rights, LAMBDA Legal Defense and Education Fund, Inc., GLBTQ Legal Advocates & Defenders, Transgender Law Center, and The Human Rights Campaign as Amicus Curiae on behalf of Defendants and Respondents.

Perkins Coie LLP and James G. Snell for Internet Association, Facebook, Inc., Glassdoor, Inc., Google LLC, and Reddit, Inc. as Amicus Curiae on behalf of Defendants and Respondents.

42