1
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
JAMES R. LAWRENCE, III (pro hac vice pending)
ANTHONY J. BILLER (pro hac vice pending)
ENVISAGE LAW
2601 Oberlin Road, Suite 100
Raleigh, North Carolina 27608
Phone: 919.755.1317
Facsimile: 919.782.0452
SEAN P. GATES (SBN 186247)
CHARIS LEX P.C.
301 N. Lake Ave., Ste. 1100
Pasadena, CA 91101
Phone: 626.508.1715
Facsimile: 626.508.1730
Attorneys for Plaintiff Alex Berenson
UNITED STATES DISTRICT COURT
NORTHERN DISTRICT OF CALIFORNIA
ALEX BERENSON,
Plaintiff,
v.
TWITTER, INC.,
Defendant.
COMPLAINT FOR DAMAGES AND
INJUNCTIVE RELIEF
1. VIOLATION OF THE FIRST
AMENDMENT
2. FEDERAL FALSE ADVERTISING
AND UNFAIR COMPETITION
3. VIOLATION OF CALIFORNIA
COMMON CARRIER LAW
4. VIOLATION OF CALIFORNIA
UNFAIR COMPETITION LAW
5. BREACH OF CONTRACT
6. PROMISSORY ESTOPPEL
7. VIOLATION OF THE
CALIFORNIA CONSTITUTION
8. UNJUST ENRICHMENT
JURY TRIAL DEMANDED
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 1 of 70
2
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
INTRODUCTION
1. Twitter is the world’s preeminent global messenger service and foremost platform
for journalism. According to its own mission statement, “defending and respecting the user’s
voice is one of our core values.” Twitter, Defending and respecting the rights of people using our
services, https://help.twitter.com/en/rules-and-policies/defending-and-respecting-our-users-voice
(last visited Dec. 19, 2021). Unfortunately, despite Twitter’s supposed commitment to freedom
of speech, in the case of independent journalist Alex Berenson, the company broke its promises,
its stated policies, and our laws, including a nearly 150-year-old California law that explicitly
sets out its obligations as a company in the business of carrying messages.
2. During 2020 and 2021, Mr. Berenson became a prominent critic of the
governmental and public health response to the COVID-19 pandemic. Mr. Berenson made many
statements on Twitter which, although controversial at the time, have since become conventional
wisdom, most notably about the profound harm to children caused by remote learning and
unnecessary school closures.
3. Mr. Berenson’s audience on Twitter soared as the pandemic continued, and in
August 2021 his tweets were viewed approximately 182 million times. Despite the controversy
around his statements, a senior Twitter executive repeatedly assured Mr. Berenson that the
company backed his right to free expression and that he would continue to enjoy access to the
platform. Even after he started critically covering the COVID-19 vaccines, in December 2020
the executive assured Mr. Berenson that his reporting “should not be an issue at all.” In March
2021, the same executive told Mr. Berenson that “your name has never come up in the
discussions around these policies [(i.e., potential censorship of Twitter users who criticized the
vaccines and other COVID-19 policies)].”
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 2 of 70
3
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
4. Relying on these assurances, and aware of Twitter’s unique size and importance
as a medium for journalism, Mr. Berenson did not seriously attempt to build alternate venues for
his reporting before late May 2021. At that point, aware that government officials were
increasingly calling for Twitter and other social media platforms to censor dissenting views on
both COVID-19 policy and the COVID-19 vaccines, he began to publish on Substack, a
newsletter platform with a far smaller audience than Twitter. But Twitter remained his primary
outlet for journalism, and between May and August 2021, he provided readers thousands of
tweets.
5. Mr. Berenson’s relationship with Twitter changed dramatically over the course of
one week in July 2021. On Sunday, July 11, Dr. Anthony Fauci, President Joe Biden’s Chief
Medical Advisor, called Mr. Berenson’s comments about COVID-19 vaccine hesitancy
“horrifying.” By Friday, President Biden himself piled on, blaming social media companies for
“killing people” on account of their failure to adequately censor content. Hours after President
Biden’s comment, Twitter locked Mr. Berenson out of his account for the first time.
6. On August 28, in the wake of even more calls for censorship from government
officials, Twitter permanently suspended Mr. Berenson from its platform, citing “repeated
violations of our COVID-19 misinformation rules.” Mr. Berenson did not violate those rules.
Twitter, on the other hand, broke its promises to Mr. Berenson as well as its policies, and
violated his rights as it served the federal government’s censorship demands.
7. Twitter and other social media companies have repeatedly fended off lawsuits like
this one using section 230 of the Communications Decency Act of 1996 or CDA. The companies
have argued—and many courts have agreed—that section 230 gives them all the legal
protections publishers enjoy but none of the obligations. Twitter has also argued that, as a private
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 3 of 70
4
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
company, it has even more fundamental rights to decide whether to publish material, rights that
spring not just from section 230 but from the First Amendment.
8. This lawsuit is not a rerun of these long-running disputes. For one thing, different
laws are at issue, among them a California law that predates the CDA by 124 years and which
limits Twitter’s right to discriminate against the speech it carries. Enacted in 1872, the law
defines any company that “offers to the public to carry persons, property, or messages” as a
“common carrier.” As the law then explains, “a common carrier must, if able to do so, accept and
carry whatever is offered to him.”
9. Courts have repeatedly applied the 1872 law to telephone companies and other
technologies that did not exist at the time it was enacted. Carrying messages for the public is, of
course, the very core of Twitter’s business. As the company itself explains, “Twitter is a social
broadcast network that enables people and organizations to publicly share brief messages
instantly around the world.” Twitter, About offensive content, https://help.twitter.com/en/safety-
and-security/offensive-tweets-and-content (last visited Dec. 19, 2021).
10. Section 230, which preempts certain state and local laws, does not shield Twitter
from the reach of California’s common carrier statute or the California Constitution. Twitter’s
potential reliance upon section 230 as a defense against Mr. Berenson’s claims asserting
California speech-protection laws raises federal First Amendment issues that federal courts must
adjudicate. If section 230 deprives Mr. Berenson of the benefits of California laws that require
Twitter to meet its obligations as a messenger service, then the First Amendment itself is at
stake, because section 230, a federal law, “is the source of the power and authority by which”
Mr. Berenson’s speech “rights are lost or sacrificed.” Ry. Emp. Dep’t v. Hanson, 351 U.S. 225,
232 (1956).
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 4 of 70
5
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
11. In fact, the Supreme Court has already signaled its awareness of these issues.
Earlier this year, the Court’s current longest-serving member, Justice Clarence Thomas, openly
asked whether companies like Twitter are in fact common carriers subject to the “general
requirement to serve all comers.” Biden v. Knight First Amend. Inst. at Columbia Univ., 141 S.
Ct. 1220, 1222 (2021) (Thomas, J., concurring).
12. Mr. Berenson also has a uniquely viable claim that Twitter acted on behalf of the
federal government in censoring and barring him from to its platform. While courts have
generally refused such “state actor” complaints against social media companies, the
extraordinarily close nexus between the July 2021 statements by senior executive branch
officials—including President Biden himself—calling for censorship by such companies and
Twitter’s corresponding immediate actions against Mr. Berenson mean that this issue merits
closer scrutiny.
13. Finally, courts have repeatedly held that even section 230 cannot and does not
protect Twitter from Mr. Berenson’s breach of contract and promissory estoppel claims, which
again do not depend merely on Twitter’s overall terms of service or broad promises about
freedom of speech but on specific, repeated representations the company made to Mr. Berenson
through a senior Twitter executive over a nearly one-year-period. Nor does section 230 bar Mr.
Berenson’s Lanham Act and California Unfair Competition Law claims premised on Twitter’s
false and misleading representations concerning his reporting on the platform.
14. This case raises significant questions about private power and the state of free
speech in America. Do our laws place any limitations on Twitter’s power to discriminate against
speakers—even as it becomes the most important outlet for journalism worldwide, in part
because its promises of unfettered free speech have attracted an audience of hundreds of millions
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 5 of 70
6
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
of users? In February 2021, in response to complaints from the Indian government, Twitter
explained in a blog post that it “exists to empower voices to be heard” and that “[w]e will
continue to advocate for the right of free expression on behalf of the people we serve.” Twitter
Safety, Updates on our response to blocking orders from the Indian Government, Feb. 10, 2021,
https://blog.twitter.com/en_in/topics/company/2020/twitters-response-indian-government. When
it banned Mr. Berenson six months later, Twitter proved otherwise. This lawsuit would be
unnecessary if the company had simply followed its own oft-repeated core principles—as well as
the general rules it created to monitor posts about COVID-19 and the specific promises it made
to Mr. Berenson. Now Twitter, as it has before, will claim that section 230 is a blanket shield
protecting it from any consequences for its behavior. But even section 230 does entirely not
exempt social media companies from the laws that govern everyone else.
PARTIES
15. Twitter is a Delaware corporation with a principal place of business located at
1355 Market Street, Suite 900, San Francisco, California 94103.
16. Alex Berenson is a resident of the State of New York.
JURISDICTION
17. Twitter is a citizen of Delaware and California. Mr. Berenson is a citizen of New
York. Twitter and Mr. Berenson have complete diverse citizenship. Mr. Berenson seeks damages
of more than $75,000. This Court has subject matter jurisdiction over this case under 28 U.S.C.
§ 1332.
18. This case presents questions that arise under the First Amendment. As such, this
Court has subject matter jurisdiction over this case under 28 U.S.C. § 1331. This Court has
supplemental jurisdiction over Mr. Berenson’s state law claims under 28 U.S.C. § 1367.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 6 of 70
7
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
19. Twitter’s headquarters are in San Francisco, California. This Court has general
personal jurisdiction over Twitter under the Due Process Clause of the Fourteenth Amendment.
See Ford Motor Co. v. Montana Eighth Jud. Dist. Ct., 141 S. Ct. 1017, 1024 (2021). This Court
also has jurisdiction over Twitter pursuant to California’s Long-Arm Statute, Cal. Code Civ.
Proc. Code § 410.10, which provides that “[a] court of this state may exercise jurisdiction on any
basis not inconsistent with the Constitution of this state or of the United States.”
VENUE
20. Venue is proper in this judicial district under 28 U.S.C. § 1391 because Twitter’s
headquarters are in this judicial district. Further, Twitter’s Terms of Service provide that “[a]ll
disputes related to these Terms or the Services will be brought solely in the federal or state courts
located in San Francisco County, California, United States.”
DIVISIONAL ASSIGNMENT
21. Pursuant to Civil L.R. 3-2(c), this case should be assigned to the San Francisco
division because a substantial portion of the events giving rise to this case occurred at Twitter’s
corporate headquarters located in San Francisco.
LEGAL FRAMEWORK
I. Twitter and the Telegraph
22. Twitter’s “social media platform . . . allows its users to electronically send
messages of limited length to the public.” Knight First Amend. Inst. at Columbia Univ. v. Trump,
928 F.3d 226, 230 (2d Cir. 2019), cert. granted, judgment vacated sub nom. Biden v. Knight
First Amend. Inst. at Columbia Univ., 141 S. Ct. 1220 (2021). A Twitter “user can post their own
messages (referred to as tweeting)” and “may also respond to the messages of others (replying),
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 7 of 70
8
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
republish the messages of others (retweeting), or convey approval or acknowledgement of
another’s message by ‘liking’ the message.” Id.
23. As one of the world’s leading social media websites, the company’s user base
numbers in the hundreds of millions. According to a 2019 survey, “[a]round one-in-five U.S.
adults say they use Twitter.” Adam Hughes & Stefan Wojcick, 10 facts about Americans and
Twitter, Pew Res. Ctr., Aug. 2, 2019, https://www.pewresearch.org/fact-tank/2019/08/02/10-
facts-about-americans-and-twitter/. “In the three months ended December 31, 2020, [Twitter]
had 192 million average [monetizable daily active users or] MDAU,” meaning the number of
“people, organizations, or other accounts who logged in or were otherwise authenticated and
accessed Twitter on any given day.” Twitter, Annual Report (Form 10-K), at 39 (Feb. 17, 2021).
The company leverages this user base to derive billions of dollars in advertising and licensing
revenues. Id. at 43.
24. The ubiquity of the company’s platform gives it significant influence over public
conversation. Among U.S. adult Twitter users, “more than half of those users (55%) get news on
the site regularly.Mason Walker & Katerina Eva Matsa, News Consumption Across Social
Media in 2021, Pew Res. Ctr., Sept. 20, 2021,
https://www.pewresearch.org/journalism/2021/09/20/news-consumption-across-social-media-in-
2021/. The demographics of Twitter users are different than the general American population,
skewing younger and toward a particular political party, see id., making Twitter a critical
platform for engaging with these fellow citizens.
25. Twitter has significant market power. According to one source, Twitter is now the
second-largest social media platform in the world, trailing only Facebook/Meta. StatCounter,
Social Media Stats Worldwide, https://gs.statcounter.com/social-media-stats (last visited Dec. 19,
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 8 of 70
9
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
2021). Twitter’s market share appears to have increased since 2018, when one report placed the
company as the third-largest platform, behind Facebook/Meta and Pinterest. Nina Angelovska,
Facebook Losing Users to Pinterest, Youtube and Twitter (Market Share By Region), Forbes
(Jan. 7, 2019), https://www.forbes.com/sites/ninaangelovska/2019/01/07/facebook-loosing-users-
to-pinterest-youtube-and-twitter-market-share-by-region/?sh=6ab17b227746.
26. Upon information and belief, these reports underestimate Twitter’s market power
in the social media messaging market, because Twitter’s platform differs significantly from the
other entities included in the comparable market in the above-cited reports. Platforms such as
Telegram are more like Twitter than Facebook/Meta is. Compared to those entities, Twitter is the
clear market leader.
27. Bolstering its market power, Twitter also boasts a significant patent portfolio,
which it can use to exclude new market entrants and competitors. Twitter Patent Portfolio – A
Little Organic and Mostly Acquired, GreyB, https://www.greyb.com/twitter-patent-portfolio/
(last visited Dec. 19, 2021). The company enjoys other incumbent advantages inherent in having
the economies of scale to comply with an ever-growing number of data security and privacy
regulations. See, e.g., Cal. Civ. Code §§ 1798.100 to 1798.199.100.
28. Twitter’s role in public debate in the twenty-first century resembles that of the
telegraph in the nineteenth. While the telegraph was initially thought to be owned and operated
by the government, private companies commercialized the telegraph, creating “the possibility
that the private entities that controlled this amazing new technology would use that power to
manipulate the flow of information to the public when doing so served their economic or
political self-interest.Genevieve Lakier, The Non-First Amendment Law of Freedom of Speech,
134 Harv. L. Rev., 2299, 2321 (2021).
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 9 of 70
10
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29. In 1872, the House of Representatives Appropriations Committee reported that
“the Government must have its ultimate guarantee in the intelligence of the people,” and that
telegraph companies “have so hedged themselves in by alliances with press associations that no
new or projected journal can have the use of the telegraph at rates not absolutely ruinous.” Id. at
2322 (quoting H.R. Rep. No. 42-6, at 7 (1872)). The House Committee proposed that the
telegraph should be subject to the “same general laws of equality of privileges” that applied to
the postal service. Id. at 2323 (quoting H.R. Rep. No. 42-6, at 8 (1872)).
30. The same year, the Senate Committee on Post-Offices and Post-Roads reported on
“[t]he danger which would result if the control of the telegraph should pass into the hands of
unscrupulous parties.” S. Rep. No. 42-20, at 4 (1872). The Committee noted the “influenc[e] [on]
public opinion and action in any important crisis [that] is possessed by those who control the
telegraph.” Id. “Such power is too vast to be confided to any private individual,” the Committee
concluded. Id. “It should be held by the Government under such restraints as would prevent its
being wielded to the public detriment.” Id.
31. Legislatures responded by requiring telegraph operators to serve all comers.
“Dozens of states enacted statutes of this sort between 1848 and the early twentieth century, as
did the federal government.Lakier, supra, at 2320. Legislators justified these “common carrier
or quasi-common carrier” statutes on the grounds that they “were necessary to safeguard the
democratic nature of the government and the vitality of the press.” Id. at 2319. “The result was
the creation, by the end of the nineteenth century, of a rich body of state and federal common
carrier law that not only promoted free speech values but did so quite intentionally in the belief
that this kind of regulation was crucial to the preservation of the democratic character of the
mass public sphere.” Id. at 2324.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 10 of 70
11
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
32. The Supreme Court has upheld similar regulations against First Amendment
challenges. As the Court put it in upholding the constitutionality of the fairness doctrine in,
“[t]he right of free speech of a broadcaster, the user of a sound truck, or any other individual
does not embrace a right to snuff out the free speech of others.Red Lion Broadcasting Co. v.
FCC, 395 U.S. 367, 387 (1969). “[I]t is idle to posit an unbridgeable First Amendment right to
broadcast comparable to the right of every individual to speak, write, or publish.” Id. at 388. The
Court explained that “[i]t would be strange if the First Amendment, aimed at protecting and
furthering communications, prevented the Government from making radio communication
possible by requiring licenses to broadcast and by limiting the number of licenses so as not to
overcrowd the spectrum.Id. at 389; see also Associated Press v. United States, 326 U.S. 1, 20
(1945) (“It would be strange . . . if the grave concern for freedom of the press which prompted
adoption of the First Amendment should be read as a command that the government was without
power to protect that freedom.”); Lakier, supra, at 2349-50 (discussing the same).
II. The Law Governing Common Carriers and Public Forums
33. The current longest-serving member of the United States Supreme Court, Justice
Clarence Thomas, recently observed that “our legal system and its British predecessor have long
subjected certain businesses, known as common carriers, to special regulations, including a
general requirement to serve all comers.” Biden, 141 S. Ct. at 1222 (Thomas, J., concurring).
After noting the historical example of the telegraph, id. at 1223, Justice Thomas observed that
“digital platforms that hold themselves out to the public resemble traditional common carriers.”
Id. at 1224. “If the analogy between common carriers and digital platforms is correct, then an
answer may arise for dissatisfied platform users who would appreciate not being blocked: laws
that restrict the platform’s right to exclude.Id. at 1225.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 11 of 70
12
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
34. Twitter’s home state of California has such a law. In 1872, the very same year the
House and Senate Committees released their above-cited reports, California enacted a law
defining the duties and obligations of common carriers. Relevant here, under the statute, “[e]very
one who offers to the public to carry persons, property, or messages, excepting only telegraphic
messages, is a common carrier of whatever he thus offers to carry.” Cal. Civ. Code § 2168
(emphasis added). (Prior to section 2168 being enacted in 1872, California courts were already
applying common carrier principles to telegraph services. See Parks v. Alta Cal. Tel. Co., 13 Cal.
422, 424 (1859).) Notably, nothing in the text of section 2168 requires that a company have
market power to be considered a common carrier.
35. The California courts have applied this law to technologies that did not exist in
1872, finding that telephone and taxicab companies are common carriers. See, e.g., Goldin v.
Pub. Utilities Comm’n, 23 Cal. 3d 638, 662, 592 P.2d 289, 304 (1979) (“A company providing
telephone services to the public is a common carrier (Civ. Code, § 2168), and as such may not
discontinue services without good cause.” (internal citations omitted)); Bezera v. Associated Oil
Co., 117 Cal. App. 139, 143, 3 P.2d 622, 623 (1931) (“The answer of the cab company, by
failure to deny, admits that it was engaged in the business of carrying passengers for hire within
and about San Francisco and that decedent was riding as a passenger for hire. This admission
would seem to bring the cab company within the definition of a common carrier found in section
2168 of the Civil Code.”).
36. As the California Supreme Court has explained, “section 2168 defines a ‘common
carrier’ carrier in expansive terms.” Gomez v. Superior Ct., 35 Cal. 4th 1125, 1130, 113 P.3d 41,
44 (2005). The Gomez court noted California’s “broad” concept of common carrier principles
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 12 of 70
13
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
such as “carrier of persons for reward” juxtaposed against the narrower view taken by courts in
other jurisdictions. See id. at 1133-36, 113 P.3d at 45-48.
37. Twitter’s services fit within section 2168’s definition of a “common carrier.”
Twitter holds itself out to the public as an entity willing to carry messages. In this regard, by the
company’s own admission, Twitter’s platform is “capable of delivering billions of messages,
including images and video, to hundreds of millions of people a day.” Twitter, Annual Report
(Form 10-K), at 8 (Feb. 17, 2021) (emphasis added). At the same time, the company understands
it faces competition from “[c]ompanies that offer products that enable people to create and share
ideas, videos, and other content and information.” Id. Twitter also says it “is an open service
that’s home to a world of diverse people, perspectives, and information,” and that “[y]ou should
be able to speak your mind and find credible information easily.” Twitter, Healthy conversations,
https://about.twitter.com/en/our-priorities/healthy-conversations (last visited Dec. 19, 2021).
38. This Court has applied section 2168 to another twenty-first century technology:
the ride-sharing application Uber. In one case, this Court rejected Uber’s argument “that it is not
a common carrier but that it is a ‘broker’ of transportation services,” Doe v. Uber Techs., Inc.,
184 F. Supp. 3d 774, 786 (N.D. Cal. 2016), holding that the plaintiff had plausibly alleged “that
Uber is a common carrier” under section 2168, id. at 787. In another case, this Court also
rejected Uber’s argument “that it is not a common carrier because of conditions that are placed
on who can use and be connected with drivers via the app.” Doe v. Uber Techs., Inc., No. 19-
CV-03310-JSC, 2019 WL 6251189, at *6 (N.D. Cal. Nov. 22, 2019). This Court noted
“[c]ommon carrier liability can attach even for transportation services that are of possible use to
only a fraction of the population.” Id. (internal quotation marks omitted). “In sum, Uber’s status
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 13 of 70
14
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
as an app-based transportation network does not preclude it as a matter of law from being held
liable as a common carrier.Id.
1
39. Under California law, common carriers like Twitter “must, if able to do so, accept
and carry whatever is offered to him, at a reasonable time and place, of a kind that he undertakes
or is accustomed to carry.” Cal. Civ. Code § 2169. In other words, since it is a common carrier
under California law, Twitter must “serve all comers.” Biden, 141 S. Ct. at 1222 (Thomas, J.,
concurring). California law goes further, requiring a “common carrier of messages . . . must
transmit messages in the order in which he receives them.” Cal. Civ. Code § 2208. “Every person
whose message is refused or postponed, contrary to the provisions of this Chapter, is entitled to
recover from the carrier his actual damages, and fifty dollars in addition thereto.” Id. § 2209
(emphasis added).
40. California case law confirms that a common carrier’s right to refuse service is
narrow. In People v. Brophy, 49 Cal. App. 2d 15, 120 P.2d 946 (1942), which involved the use of
telephone services in connection with a bookmaking operation, the court stated that “common
carriers are not the censor of public or private morals.49 Cal. App. 2d at 33. As the Brophy
court put it, a “telephone company has no more right to refuse its facilities to persons because of
a belief that such persons will use such service to transmit information that may enable recipients
thereof to violate the law than a railroad company would have to refuse to carry persons on its
trains because those in charge of the train believed that the purpose of the persons so transported
in going to a certain point was to commit an offense.” Id.; cf. Mason v. Western Union Telegraph
1
This Court ultimately rejected the plaintiff’s common carrier theory because she had not
alleged facts sufficient to establish “a common carrier/passenger relationship” between her and
Uber. Id. at *8.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 14 of 70
15
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Co., 52 Cal.App.3d 429, 439, 125 Cal. Rptr. 53, 58 (1975) (“We conclude that Western Union
has no right of censorship except in the specific instances enumerated in Public Utilities Code
section 7904 which instances do not include the right to refuse to transmit telegrams which may
allegedly be libelous.”). If common carriers cannot bar even libelous material, certainly speech
about matters of public concern does not fall within such a carrier’s narrow right to exclude. Of
course, this case involves the narrow issue of Twitter’s right to exclude Mr. Berenson.
41. As a speech-protective statute, section 2168 supplements the rights guaranteed by
the California Constitution. The California Supreme Court has interpreted “sections 2 and 3 of
article I of the California Constitution [to] protect speech and petitioning, reasonably exercised,
in shopping centers even when the centers are privately owned.” Robins v. Pruneyard Shopping
Ctr., 23 Cal. 3d 899, 910, 592 P.2d 341, 347 (1979). In particular, the California Supreme Court
noted that the shopping center was not “an individual homeowner or proprietor,” but that “[a]s a
result of advertising and the lure of a congenial environment, 25,000 persons are induced to
congregate daily to take advantage of the numerous amenities offered” there. Id. at 911, 592 P.2d
at 347. The Supreme Court later affirmed Pruneyard against a First Amendment challenge by the
shopping center. Pruneyard Shopping Ctr. v. Robins, 447 U.S. 74 (1980).
42. Pruneyard’s logic, and the speech protections that case brings with it, extend to
Twitter. As a publicly traded company with a market capitalization greater than $40 billion,
Twitter is hardly “an individual homeowner or proprietor.” The company induces hundreds of
millions of people to use its platform daily. And unlike the shopping center in Pruneyard,
Twitter’s business model revolves around speech—around facilitating a global conversation.
43. Extending Pruneyard to Twitter, or applying section 2168 to the company, would
not violate the First Amendment. In Pruneyard, the Supreme Court rejected the shopping
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 15 of 70
16
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
center’s argument “that a private property owner has a First Amendment right not to be forced by
the State to use [its] property as a forum for the speech of others.” Pruneyard, 447 U.S. at 85.
Distinguishing the case from Wooley v. Maynard, 430 U.S. 705 (1977), in which the Supreme
Court struck down New Hampshire’s mandatory display of “Live Free or Die” on non-
commercial license plates, the Court noted the shopping center “is not limited to personal use,”
but is rather “a business establishment that is open to the public to come and go as they please.”
Pruneyard, 477 U.S. at 87. As noted above, the same is true of Twitter.
44. The Pruneyard Court further explained that “[t]he views expressed by members
of the public in passing out pamphlets or seeking signatures for a petition thus will not likely be
identified with those of the owner.” Id. The shopping center owner “can expressly disavow any
connection with the message by simply posting signs in the area where the speakers or
handbillers stand.” Id.; see also Biden, 141 S. Ct. at 1224 (Thomas, J., concurring) (noting
“regulations that might affect speech are valid if they would have been permissible at the time of
the founding,” and that would be particularly true “where a restriction would not prohibit the
company from speaking or force the company to endorse the speech”). Twitter is of course free
to disavow speech on its platform and often does.
45. Nor does the Dormant Commerce Clause provide Twitter with the ability to evade
section 2168 or Pruneyard. The Dormant Commerce Clause “holds that any ‘statute that directly
controls commerce occurring wholly outside the boundaries of a State exceeds the inherent limits
of the enacting State’s authority.’Rocky Mountain Farmers Union v. Corey, 730 F.3d 1070,
1101 (9th Cir. 2013) (quoting Healy v. Beer Inst., 491 U.S. 324, 336 (1989)).
46. In Rocky Mountain Farmers Union, the Ninth Circuit upheld California’s fuel
standards against a Dormant Commerce Clause challenge, finding the standards were based “on
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 16 of 70
17
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
the harmful properties of fuel” and the standards “do[] not control the production or sale of
ethanol wholly outside of California.” Id. at 1104. “States may not mandate compliance with
their preferred policies in wholly out-of-state transactions, but they are free to regulate commerce
and contracts within their boundaries with the goal of influencing the out-of-state choices of
market participants.Id. at 1103. Neither section 2168 nor the California Constitution regulate
conduct outside of California since Twitter’s censorship decisions emanate from its California
headquarters.
III. Section 230 of the Communications Decency Act
47. Twenty-five years ago, Congress enacted the Telecommunications Act of 1996,
Pub. L. No. 104-104, 110 Stat. 56 (Feb. 8, 1996). Within that statute lies the Communications
Decency Act of 1996 or CDA. The CDA is Title V of the Telecommunications Act of 1996 with
the heading “Obscenity and Violence.” 110 Stat. at 133. Section 230 of the CDA is in Subtitle A
of Title VII, which Congress captioned “Obscene, Harassing, and Wrongful Utilization of
Telecommunications Facilities,” id., tucked alongside requirements to scramble “sexually
explicit adult programming,” id. at 136. Congress placed section 230 of the CDA in section 509
of the Telecommunications Act of 1996 under the family-first heading “Online Family
Empowerment.” Id. at 137.
48. As one of the architects of section 230 explained at the time, the Internet contains
“some offensive material, some things in the bookstore, if you will, that our children ought not to
see.” 141 Cong. Rec. H8469 (statement by Rep. Cox). Noting that “the existing legal system
provides a massive disincentive for the people who might best control the Internet to do so,” id.,
namely the litigation risk for some Internet platforms for policing any content at all,
Representative Cox explained section 230 would “protect computer Good Samaritans . . . who
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 17 of 70
18
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
take steps to screen indecency and offensive materials for their customers,” and “keep[]
pornography away from our kids,” id. at H8470. Then-Representative Wyden, section 230’s co-
sponsor, similarly argued that “parents and families are better suited to guard the portals of
cyberspace and protect our children than our Government bureaucrats.” Id.
49. Another supporter of section 230 noted that statute “allows parents to make
important decisions with regard to what their children can access, not government.” Id. at H8472
(statement by Rep. Goodlatte). Critically, Representative Goodlatte stated the law “doesn’t
violate free speech or the right of adults to communicate with each other.” Id.
50. The operative provision of section 230 is subsection (c):
(c) PROTECTION FOR “GOOD SAMARITAN BLOCKING AND SCREENING OF OFFENSIVE
MATERIAL
(1) TREATMENT OF PUBLISHER OR SPEAKER
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.
(2) CIVIL LIABILITY
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; or
(B) any action taken to enable or make available to information
content providers or others the technical means to restrict access to
material described in paragraph (1).
47 U.S.C. § 230(c).
51. Section 230 further provides that “[n]o cause of action may be brought and no
liability may be imposed under any State or local law that is inconsistent with this section.Id.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 18 of 70
19
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
§ 230(e)(3). But elsewhere in the Telecommunications Act of 1996, Congress provided that
“[t]his Act and the amendments made by this Act [(i.e., section 230)] shall not be construed to
modify, impair, or supersede Federal, State, or local law unless expressly so provided in such
Act or amendments.” Telecommunications Act of 1996, § 601(c)(1), 110 Stat. at 143.
52. Social media companies like Twitter contend that section 230 effectively nullifies
speech-protective state laws, allowing the platforms to censor otherwise constitutionally
protected speech. That is an odd outcome given Congress’ focus on pornography in the CDA.
Congress even explicitly found that “[t]he Internet and other interactive computer services offer
a forum for a true diversity of political discourse, unique opportunities for cultural development,
and myriad avenues for intellectual activity.” 47 U.S.C. § 230(a)(3). Interposing section 230 to
nullify California’s common carrier statute and the free speech and petition guarantees of the
California Constitution to facilitate censoring speech about matters of public concern undermines
the very free expression Congress said it wanted to encourage, invading the sovereign
prerogatives of the States at the same time.
53. Courts have read section 230 to immunize Internet platforms from liability where
plaintiffs sued them for their publishing decisions. See, e.g., Barnes v. Yahoo!, Inc., 570 F.3d
1096, 1100-01 (9th Cir. 2009) (holding that section 230(c)(1) “protects 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”); Fair Housing Council of San Fernando Valley v.
Roommates.Com, 521 F.3d 1157, 1170-71 (9th Cir. 2008) (en banc) (“And 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.”).
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 19 of 70
20
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
54. Plaintiffs in those cases did not raise the California common carrier statute, Cal.
Civ. Code § 2168, or otherwise seek to treat the platforms as a common carrier as opposed to a
“publisher or speaker” in section 230(c)(1)’s parlance. Tellingly, Congress addressed the
applicability of common carriage principles to “interactive computer services” in section 502 of
the CDA, 47 U.S.C. § 223, providing that “[n]othing in this section,” which addressed obscene
or harassing phone calls, “shall be construed to treat interactive computer services as common
carriers or telecommunications carrier.” 110 Stat at 135 (codified at 47 U.S.C. § 223(e)(6)).
California’s common carrier statute is not “in this section,” i.e., the California law is not in 47
U.S.C. § 223. If Congress intended to bar treatment of platforms as common carriers, it could
have explicitly done so, but did not, which is telling considering the history of common carriage
regulation by the States. Lakier, supra, at 2320 (noting that “[d]ozens of states enacted [common
carriage] statutes of this sort between 1848 and the early twentieth century”). If Congress had
wanted to bar the possibility of applying these laws to the Internet, it did not do so with
“exceedingly clear language,” even though an expansive reading of section 230 would
significantly alter the ability of States like California to govern themselves in this area. Cf.
United States Forest Serv. v. Cowpasture River Pres. Ass’n, 140 S. Ct. 1837, 1849–50 (2020)
(“Our precedents require Congress to enact exceedingly clear language if it wishes to
significantly alter the balance between federal and state power and the power of the Government
over private property.”).
55. Moreover, section 230 does not categorically bar breach of contract, promissory
estoppel, and least some commercial tort claims. See Barnes, 570 F.3d at 1109 (holding that
“subsection 230(c)(1) . . . does not preclude” a claim “alleg[ing] breach of contract . . . under the
theory of promissory estoppel”); Enigma Software Grp. USA, LLC v. Malwarebytes, Inc., 946
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 20 of 70
21
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
F.3d 1040, 1054 (9th Cir. 2019), cert. denied, 141 S. Ct. 13, 208 L. Ed. 2d 197 (2020) (holding
that “immunity under [section 230] does not extend to anticompetitive conduct”).
56. Ultimately, an expansive, pro-censorship interpretation of section 230,
particularly one that disregards speech-protective state laws, presents First Amendment
challenges of its own. The First Amendment protects freedom of association. Roberts v. U.S.
Jaycees, 468 U.S. 609, 622 (1984) (“An individual’s freedom to speak, to worship, and to
petition the government for the redress of grievances could not be vigorously protected from
interference by the State unless a correlative freedom to engage in group effort toward those ends
were not also guaranteed.”). In Railway Employees Department v. Hanson, 351 U.S. 225 (1956),
the Supreme Court considered the interplay between a provision of the Nebraska Constitution
which prohibited employers from “exclud[ing] persons from employment because of
membership in or nonmembership in a labor organization” and a federal statute that empowered
employers to “requir[e] all employees within a stated time to become a member of the labor
organization.” Id. at 228.
57. From the employee’s perspective, the Nebraska Constitution granted associational
freedoms beyond those guaranteed by the First Amendment, namely associational freedoms
enforceable against private employers. Even though “private rights” were at issue, the Supreme
Court determined the case raised “justiciable questions” under the First Amendment because
“[i]f private rights are being invaded, it is by force of an agreement made pursuant to federal law
which expressly declares that state law is superseded.” Id. at 232. In this regard, the Supreme
Court approvingly quoted the Supreme Court of Nebraska’s observation that “[s]uch action on
the part of Congress is a necessary part of every union shop contract entered into on the railroads
as far as these 17 states are concerned for without it such contracts could not be enforced
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 21 of 70
22
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
therein,id. (quoting Hanson v. Union Pac. R. Co., 160 Neb. 669, 698, 71 N.W.2d 526, 547
(Neb. 1955)), before ultimately determining the federal statute did not violate the First
Amendment, id. at 238.
58. California’s speech-protective laws and section 230 raise First Amendment issues
under Hanson. Like the right to work provision in the Nebraska Constitution at issue in that case,
both the California common carrier statute and the California Constitution provide more
protection for speech than the First Amendment in the abstract because they grant rights against
private parties. When an “interactive computer service” subject to these state laws censors
speech in violation of those laws, it is “by force of” section 230, “a federal law which expressly
declares that state law is superseded.” Id. Federal law is the source of the private censorship, just
like the federal statute in Hanson was the source of the limits on employee freedom of
association, raising “justiciable questions” under the First Amendment.
59. For independent journalist and best-selling author Alex Berenson, all this was
largely academic until July 2021. As detailed below, during Mr. Berenson’s decade-plus on
Twitter the company largely lived up to its stated commitments to free speech, furnishing him a
platform to report on numerous issues, including the COVID-19 pandemic. But then, this past
summer, something changed. Despite the company’s prior assurances, and just hours after the
President of the United States accused social media platforms of “killing people” for not
censoring speech about COVID-19 vaccines, the company locked Mr. Berenson out of his
account. A month later, it permanently suspended him. How did it come to this?
///
///
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 22 of 70
23
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
FACTUAL BACKGROUND
I. Alex Berenson’s Use of Twitter Before and During the Early Days of the COVID-19
Pandemic
60. Throughout his career, and like many other journalists, Mr. Berenson has used
social media to amplify his reporting. Mr. Berenson joined Twitter in November 2009. The
company verified his account in or around 2014, affixing a blue check mark to his account, a
sign it was “authentic, notable, and active.” Twitter, About Verified Accounts,
https://help.twitter.com/en/managing-your-account/about-twitter-verified-accounts (last visited
Dec. 19, 2021). As of January 2020, Mr. Berenson had approximately 7,000 Twitter followers,
many of whom were drawn to his previous books and reporting.
61. In January 2020, Mr. Berenson tweeted for the first time about COVID-19. In that
tweet, he shared an article about SARS-CoV-2, the coronavirus that causes COVID-19, by New
York Times columnist Nicholas Kristof. The tweet attracted one “like.” At the time, much of the
press did not appreciate that the COVID-19 pandemic would become, as NBC’s Lester Holt put
it later, “the story of my lifetime.” Ted Johnson, NBC News’ Lester Holt On Reporting On The
Coronavirus: “We’ve All Covered The Burning House, But This Time We’re In The House” –
Q&A, Deadline (Mar. 18, 2020), https://deadline.com/2020/03/coronavirus-lester-holt-nbc-news-
q-and-a-qa-1202886983/.
62. Mr. Berenson dedicated much of his career to reporting on and breaking
important news stories. Beginning in March 2020, his attention increasingly turned to COVID-
19. This was as governments around the world, including state and local governments here in the
United States, issued far-reaching, and in many ways unprecedented, lockdown orders as part of
an effort “flatten the curve” of viral spread and transmission.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 23 of 70
24
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
63. In late March 2020, Mr. Berenson reported on revised epidemiology models
published by Professor Neil Ferguson of Imperial College London. In a six-tweet thread
published on March 26, 2020, Mr. Berenson discussed Dr. Ferguson’s revised mortality
estimates, calling them a “remarkable turn” for an academic who had earlier forecast that
500,000 people would die from COVID-19 in the United Kingdom.
64. Mr. Berenson’s March 26 tweet was his first to go “viral.” Industrialist and
entrepreneur Elon Musk, among other prominent figures, retweeted Mr. Berenson’s post. The
thread’s initial tweet, which is shown below, drew 7,468 likes, 3,743 retweets, and more than 4.6
million impressions. Twitter did not label this tweet false or misleading.
65. In April 2020, Mr. Berenson continued covering epidemiological models.
Specifically, he compared actual hospitalization data in New York and other jurisdictions against
projections from the Institute for Health Metrics and Evaluation or IMHE. On April 6, New York
Times White House correspondent Maggie Haberman noted Mr. Berenson’s tweets “are making
their way around the White House among some senior officials.” Twitter did not label any of Mr.
Berenson’s reporting false or misleading.
66. During the month of April 2020, Mr. Berenson’s Twitter presence grew
exponentially. He added more than 77,000 followers and his tweets from that month garnered
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 24 of 70
25
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
175 million impressions. As of the start of May, Mr. Berenson had increased his Twitter
following by fourteen-fold from its early March 2020 mark, up from 7,000 to 84,000 followers.
67. Mr. Berenson’s reporting drew criticisms, with some calling for his voice to be
silenced. At the same time, his work drew support from others across America and around the
world who were suffering under lockdown policies.
II. Twitter’s Evolving Approach to Medical Misinformation During the Early Days of
the COVID-19 Pandemic
68. In early March 2020, Twitter began to set out its approach to COVID-19. In a
March 4, 2020 blog post, the company pledged its support for “[p]rotecting the conversation.”
The company explained that commitment as follows: “The power of a uniquely open service
during a public health emergency is clear. The speed and borderless nature of Twitter presents an
extraordinary opportunity to get the word out and ensure people have access to the latest
information from expert sources around the world.” Twitter Public Policy, Stepping up our work
to protect the public conversation around Covid-19, (Mar. 4, 2020),
https://blog.twitter.com/en_us/topics/company/2020/stepping-up-our-work-to-protect-the-public-
conversation-around-covid-19.
69. The company also expressed its desire to “help people find authoritative health
information.” Id. “With a critical mass of expert organizations, official government accounts,
health professionals, and epidemiologists on our service,” Twitter explained, “our goal is to
elevate and amplify authoritative health information as far as possible.Id. Twitter did not define
“authoritative health information” in the post.
70. Twitter stated that it “put additional safeguards in place in order to facilitate the
sharing of trusted public health information and to reduce potential harm to users,an effort
largely focused on “advertisers [who might] opportunistically use the Covid-19 outbreak to
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 25 of 70
26
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
target inappropriate ads.” Id. The company did not forecast censorship. While Twitter
encouraged users to “[f]ollow @WHO [(the World Health Organization’s Twitter account)] and
your local health ministry” and “seek out authoritative health information,” the company advised
users to “ignore the noise.Id.
71. In a separate blog post published on March 16, 2020, which the company updated
twelve days later on April 1, Twitter unveiled its approach to content moderation. The company
announced it was “[b]roadening our definition of harm to address content that goes directly
against guidance from authoritative sources of global and local public health information.”
@Vijaya & Matt Derella, An update on our continuity strategy during COVID-19, Twitter (Mar.
16, 2020, updated Apr. 1, 2020), https://blog.twitter.com/en_us/topics/company/2020/An-
update-on-our-continuity-strategy-during-COVID-19.
72. Again, the blog post did not elaborate on what the company considered to be
“authoritative sources.” Twitter stated it would “enforce this [policy] in close coordination with
trusted partners, including public health authorities and governments, and continue to use and
consult with information from those sources when reviewing content.Id. (emphasis added).
Even so, the company explained it did not intend “to limit good faith discussion or expressing
hope about ongoing studies related to potential medical interventions that show promise.Id. As
of March 18, 2020, Twitter reported it had “removed more than 1,100 tweets containing
misleading and potentially harmful content from Twitter.” Id. At that point, Twitter had not
labeled or removed any of Mr. Berenson’s tweets.
73. The post further identified eleven categories of information Twitter “will require
people to remove.” Id. The post targeted obvious falsehoods, including claims such as “the news
about washing your hands is propaganda for soap companies, stop washing your hands.” Id. For
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 26 of 70
27
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
other claims, as detailed below, Twitter distinguished between factual assertions, opinions, and
specific calls to action or inaction to other Twitter users.
a. Twitter called for censoring the “[d]enial of global or local health
authority recommendations to decrease someone’s likelihood of exposure to COVID-19,” but
required such statements be made “with the intent to influence people into acting against
recommended guidance.” Id. (emphasis added). The company cited the claim “social distancing
is not effective” as an example but tied that remark to “actively encouraging people to not
socially distance themselves in areas known to be impacted by COVID-19 where such measures
have been recommended by the relevant authorities.” Id.
b. Twitter stated it would take down discussion of “alleged cures” and
“harmful treatments or protection measures” that “are known to be ineffective.” Id. The company
cited “use aromatherapy and essential oils to cure COVID-19” as a claim that could cause harm
by inducing Twitter user to delay or forgo needed medical treatment. The other example,
“drinking bleach and ingesting colloidal silver will cure COVID-19,” was so obviously false that
that the company noted such claims would be taken down “even if made in jest.” Notably,
Twitter said nothing about discussion or debate concerning clinical data, government
information, or research articles regarding non-pharmaceutical interventions or vaccines, drugs,
or other therapeutics.
c. The company said it would censor “false or misleading information
around COVID-19 diagnostic criteria or procedures,” and “[f]alse or misleading claims on how
to differentiate between COVID-19 and a different disease.” Again, the company targeted
content aimed at influencing behavior, which in turn could delay or keep an individual from
seeking treatment, not discussion or debate about policy or issues.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 27 of 70
28
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
74. On May 11, 2020, Twitter announced further action on COVID-19 information.
In another blog post, the company explained it “may use labels and warning messages to provide
additional explanations or clarifications in situations.” Yoel Roth & Nick Prickles, Updating our
approach to misleading information, Twitter (May 11, 2020),
https://blog.twitter.com/en_us/topics/product/2020/updating-our-approach-to-misleading-
information.
75. The company defined “three broad categories” of “false or misleading content”:
Misleading information — statements or assertions that have been
confirmed to be false or misleading by subject-matter experts, such as
public health authorities.
Disputed claims — statements or assertions in which the accuracy,
truthfulness, or credibility of the claim is contested or unknown.
Unverified claims — information (which could be true or false) that is
unconfirmed at the time it is shared.
Id. The company did not identify any of the “subject-matter experts” or “public health
authorities” it considered to be “authoritative sources.”
76. Twitter explained that “disputed claims” and “misleading information” with a
“moderate” risk of harm were subject to being labeled with a “link to a Twitter-curated page or
external source containing additional information on the claims made within the Tweet.” Id. The
company gave the following as an example of labeled false content:
///
///
///
///
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 28 of 70
29
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
77. Twitter explained that “misleading information” with a propensity to cause
“severe” harm would be removed. Id. Meanwhile, a disputed claim that might cause severe harm
would be labeled with a “warning.” Twitter gave the following as an example:
78. As with its previous COVID-19 policies, nowhere in the blog post did Twitter bar
discussion or analysis of clinical data, government information, or research articles. In a
contemporaneous news report on the policy, a Twitter official explained that “[p]eople don’t
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 29 of 70
30
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
want us to play the role of deciding for them what’s true and what’s not true but they do not want
people to play a much stronger role providing context.” Amanda Seitz, Twitter to label disputed
COVID-19 tweets, AP News, May 11, 2020, https://apnews.com/article/virus-outbreak-health-us-
news-ap-top-news-technology-c8a542e2f22004c0c06cbbe1e1b58a52. To be clear, at least at this
point, and viewed in the context of Twitter’s prior instruction to its users to “ignore the noise,”
the company was plotting a course for more speech, not less.
III. With Twitter’s full knowledge and personal assurances, Mr. Berenson continues to
cover COVID-19 on Twitter, and his following grows.
79. Mr. Berenson covered COVID-19 throughout March, April, and May 2020.
Among other things, during and after that time, Mr. Berenson used his Twitter account to
repeatedly criticize universal masking policies, without regard for whether an individual is sick
or well, as an effective COVID-19 preventative measure and other non-pharmaceutical
interventions such as business and school closures.
80. On the same day Twitter announced its new labeling policy, and though none of
his tweets had been taken down or modified at the time, Mr. Berenson tweeted out his concern
that the company was going to start censoring content. A few hours after Mr. Berenson’s tweet,
Twitter then-CEO Jack Dorsey followed his account. At that point, Twitter’s leadership at the
highest level was fully aware of Mr. Berenson’s reporting.
81. The same day Mr. Dorsey followed Mr. Berenson, Brandon Borrman, who then
served as Twitter’s Vice President of Global Communications, contacted Mr. Berenson about
censorship issues. “I work at Twitter and saw your Tweets today,” Mr. Borrman said. “Would
you be open to having a discussion so I can hear you out? I think you have some nuanced points
that could be helpful as [sic] try to move ahead.”
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 30 of 70
31
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
82. Mr. Berenson responded, acknowledging Twitter’s legitimate interest in blocking
claims “that the virus is not real, or that it’s part of a UN conspiracy to sterilize America, or
similar nonsense.” Mr. Berenson summarized his approach to reporting on COVID-19 as
follows:
I am trying to raise serious, data-driven questions, based wherever possible on
government data or peer-reviewed/preprint papers—I think my most recent tweets
tonight capture the flavor. Does universal masking work as a broad policy mandate?
The truth is the evidence is pretty weak—not that sick people shouldn’t wear them,
or maybe that people in confined public transportation (aka NY subways)
shouldn’t, but what states like NY and CA are saying and doing right now far
outruns the evidence.
83. Twitter’s communications executive responded the following day. “We are trying
to take a more nuanced approach to this that recognizes that there is a huge amount of emotion
and vitriol [on] all sides of the issue,” Mr. Borrman explained. “We’re trying to make sure that
factual debate finds a way through the emotion, but we’re obviously not successful all the time.”
84. Mr. Borrman explained that “[t]he mask example was a poor one to use.”
2
Appearing to draw on Twitter’s prior policies which recognized a distinction between debate
about policy and calls to violate specific laws, Mr. Borrman explained that
We’re less likely to take action on something regarding the efficacy of masks, and
more likely to label something that, for example, directly calls on people to violate
a local order to wear masks, regardless of their efficacy. In other words, we’re
trying to avoid appearing as though we’re saying one treatment or solution is good
or bad while still recognizing that, even if controversial, there are new laws in place
in many locations. Debate over the effectiveness of those laws should continue, but
encouragements to directly violate them cross a line.
2
Upon information and belief, Mr. Borrman was referring to the statement in the above-cited
Associated Press article that “[t]he company has been removing bogus coronavirus cures and
claims that social distancing or face masks do not curb the virus’ spread for several weeks.”
Seitz, supra.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 31 of 70
32
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
As Mr. Borrman put it, Twitter is “leading with labeling here, rather than simply deciding
between leave up/take down, because we think it will allow the conversation to continue.” While
the company “will be putting our thumbs on the scale with a label” where there are “pushes to
violate local laws,” Mr. Borrman said, “I don’t expect that you will see major change in how we
are addressing things going forward, but if you do, please feel free to reach out to me.”
85. In this e-mail exchange, Mr. Borrman did not raise specific concerns regarding
Mr. Berenson’s reporting. Instead, he advised Mr. Berenson there was no “major change”
coming. While Mr. Berenson certainly criticized the laws Twitter’s executive referred to in the e-
mail, and reported on the underlying evidentiary basis for them, at no relevant time did he
encourage individuals to break those laws.
86. As a sophisticated commercial entity, Twitter employs counsel and Mr. Borrman
had ample time and opportunity to seek out legal counsel to vet the statements and assurances the
company made to Mr. Berenson. Upon information and belief, Mr. Borrman sought such counsel
and made the statements in the May 12 e-mail message with the benefit of that counsel.
87. Consistent with Mr. Borrman’s assurances, Twitter’s treatment of Mr. Berenson
stayed the same. Aside from corresponding with Mr. Borrman regarding a “dox” threat, Mr.
Berenson had no formal communication with the company. He had no reason to. Upon
information and belief, Twitter did not label any of Mr. Berenson’s tweets at any point in 2020.
88. Relying on Twitter’s assurances that no “major changes” in its moderation
practices were upcoming, Mr. Berenson continued to build his Twitter following, spending hours
each day on the platform bringing COVID-19 information to audience. Mr. Berenson invested
this time in Twitter at the opportunity cost of investing his time in building a presence on
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 32 of 70
33
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Telegram and Substack, two alternative platforms to Twitter available to him at the time, or
writing books and articles for publication elsewhere.
89. In June 2020, Mr. Berenson published his first “Unreported Truths” booklet on
COVID-19 and lockdowns. After Amazon initially blocked the pamphlet’s distribution on its
platform, Mr. Berenson brought the issue to his Twitter followers. In what became a national
news story, Elon Musk criticized Amazon for the decision, tweeting “[t]ime to break Amazon
up.” Todd Haselton, Elon Musk calls for Amazon breakup after Covid-19 skeptic claims it
censored his book, CNBC (June 4, 2020), https://www.cnbc.com/2020/06/04/elon-musk-calls-
for-amazon-split-after-alex-berenson-claims-censorship.html. Ultimately, because of this
pressure, Amazon reversed course and let Mr. Berenson sell the pamphlets on its platform.
90. Upon information and belief, Twitter was fully aware of Mr. Berenson’s conflict
with Amazon. The company was also aware of the content of Mr. Berenson’s pamphlets, which
presented data and analysis undermining the case for government lockdown measures. Mr.
Berenson sold more than 300,000 copies of his “Unreported Truths” pamphlets, and he used to
Twitter to generate much of the awareness of those works.
91. Meanwhile, back on Twitter, Mr. Berenson continued to criticize government
lockdowns and mask mandates. In an August 25, 2020 tweet, which is shown below, Mr.
Berenson said that “lockdowns have provably failed” and “masks . . . are even more useless.”
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 33 of 70
34
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
This tweet received more than 1,250 likes and retweeted more than 350 times.
92. Later, in November 2020, Mr. Berenson slammed not only lockdowns and mask
mandates, but other non-pharmaceutical interventions such as contact tracing, saying advice
from public health experts had “proven useless.” That tweet, which garnered more than 4,700
likes and 1,000 retweets, is shown below.
93. Twitter did not label any of these tweets, or any of Mr. Berenson’s other similar
tweets, as false or misleading. The company continued to allow Mr. Berenson to report freely
and without interference. All of that was consistent with Twitter’s specific, individualized prior
assurances and representations via Mr. Borrman that “[d]ebate over the effectiveness of those
laws should continue,” and that Mr. Berenson would not “see major change in how we are
addressing things going forward.”
IV. With the advent of COVID-19 vaccines, Twitter announces new policies on
misleading medical information, and the company gives further assurances to Mr.
Berenson regarding his access to the platform.
94. On November 9, 2020, Pfizer and BioNTech announced positive clinical trial
results regarding their COVID-19 vaccine. Far from being “anti-vax,” having taken vaccines
himself and allowing his children to be vaccinated, Mr. Berenson tweeted that while “more
safety data” is needed, “this is legitimately good news,” arguing that the emergence of a COVID-
19 vaccine might offer a pathway out of the pandemic once and for all.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 34 of 70
35
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
95. One week later, on November 16, Moderna released results about its vaccine
candidate. Mr. Berenson was again positive, tweeting that the development was “[m]ore good
topline vaccine news,” while linking to an article in the Washington Post with the headline
“Moderna’s coronavirus vaccine found to be nearly 95 percent effective in a preliminary
analysis.”
96. The primary focus of Mr. Berenson’s reporting throughout the rest of November
and into early December 2020 continued to be other COVID-19 issues, and not vaccines. On
December 11, Mr. Berenson tweeted about alleged political pressure related to FDA’s
authorization of the Pfizer-BioNTech COVID-19 vaccine. Three days later, on December 14,
Mr. Berenson tweeted about an article published in The New England Journal of Medicine
regarding the rate of occurrence of “severe adverse events” in a Pfizer-BioNTech vaccine trial.
97. Two days later, Twitter announced a new policy regarding misleading COVID-19
vaccine information. Twitter Safety, COVID-19: Our approach to misleading vaccine
information, Twitter (Dec. 16, 2020),
https://blog.twitter.com/en_us/topics/company/2020/covid19-vaccine. The company announced
it would start to “remove Tweets which advance harmful or misleading narratives about COVID-
19 vaccinations, including” the following:
False claims that suggest immunizations and vaccines are used to
intentionally cause harm to or control populations, including statements
about vaccines that invoke a deliberate conspiracy;
False claims which have been widely debunked about the adverse impacts
or effects of receiving vaccinations; or
False claims that COVID-19 is not real or not serious, and therefore that
vaccinations are unnecessary.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 35 of 70
36
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Id. Twitter also stated “we may label or place a warning on Tweets that advance unsubstantiated
rumors, disputed claims, as well as incomplete or out-of-context information.” Id.
98. The same day, Twitter also announced a larger “COVID-19 misleading
information policy.” COVID-19 misleading information policy, Twitter (Dec. 16, 2020),
https://web.archive.org/web/20201216200114/https://help.twitter.com/en/rules-and-
policies/medical-misinformation-policy. The company stated that for content to violate the
policy “it must” do the following: (1) “advance a claim of fact, expressed in definitive terms; (2)
“be demonstrably false or misleading, based on widely available, authoritative sources”; and (3)
“be likely to impact public safety or cause serious harm.” Id.
99. The policy set forth five categories of misleading information. The company said
it “will label or remove false or misleading about,” among other things, “official regulations,
restrictions, or exemptions pertaining to health advisories,” regarding “[p]ersonal protective
equipment (PPE) such as claims about the efficacy and safety of face masks to reduce viral
spread.” Id.
100. At the same time, Twitter said “[w]e seek to protect robust, public debate about
the response to COVID-19, and recognize that the state of scientific knowledge about certain
aspects of the pandemic and public response to it (including the development of vaccines) is still
relatively nascent.Id. To that end, Twitter specifically stated that in the absence of other policy
violations, the following are generally not in violation of this policy:
Strong commentary, opinions, and/or satire, provided these do not
contain false or misleading assertions of fact.
Counterspeech. We allow for direct responses to misleading information
which seek to undermine its impact by correcting the record, amplifying
credible information, and educating the wider community about the
prevalence and dynamics of misleading information.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 36 of 70
37
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Personal anecdotes or first-person accounts.
Public debate about the advancement of COVID-19 science and
research, including debate about research related to COVID-19, such
as the effectiveness of treatments and mitigation measures, so long as the
claims don’t misrepresent research findings.
Id.
101. In the December 16 policy, Twitter also explained that “we are enforcing this
policy in close coordination with trusted partners, including public health authorities, NGOs and
governments, and continue to use and consult with information from those sources when
reviewing content.Id. (emphasis added). Upon information and belief, Twitter’s “trusted
partners” include agencies of both state and federal governments.
102. In view of these new policies, Mr. Berenson contacted Twitter executive Brandon
Borrman on December 17, looking for assurances that Twitter would continue to allow his
reporting on COVID-19 vaccines. Many “people [are] complaining I’m raising questions about
the vaccine,” Mr. Berenson wrote to Mr. Borrman. “So you know, there’s no conspiracy theory
nonsense,” rather “[e]verything I point to about safety comes from the clinical trial data,” and
“my broader point is simply that we shouldn’t be mandating this for adults.”
103. Mr. Borrman responded less than four hours later. “The points you’re raising
should not be an issue at all,” Mr. Borrman assured Mr. Berenson. “The policy is designed to
allow debate and discussion, but to discourage conspiracy theories, etc. Please let me know if
you run into any issues.”
104. Mr. Berenson tweeted the e-mail exchange to his followers, and lauded Twitter’s
commitment to free speech. Given Twitter’s assurances, Mr. Berenson continued covering
COVID-19 on Twitter, again forgoing the opportunity to build his following and presence on
alternative platforms such as Telegram and Substack. Upon information and belief, at the same
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 37 of 70
38
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
time, Twitter continued to profit from Mr. Berenson’s coverage of COVID-19 by monetizing his
following through the sale of advertisements.
105. Mr. Berenson kept on tweeting about mandatory mask policies and lockdowns.
On January 28, 2021, Mr. Berenson criticized the media “for refusing to report the real-world
evidence” that masks and lockdowns are “useless.” Several weeks later, Mr. Berenson again
called mask “useless.” With respect to vaccines, Mr. Berenson raised questions about side effects
and the long-term efficacy of the shots. Mr. Berenson often linked to primary and secondary
source materials from which he was drawing these conclusions.
106. Twitter did not label any of this reporting false or misleading. Nor did the
company take any of the posts down. Given Twitter’s personal assurances to him, Mr. Berenson
had no reason to believe the company would take action against his account or keep him from
reporting on Twitter.
V. Twitter introduces its new five-strike policy, reassures Mr. Berenson, and declines
to “strike” any of his subsequent tweets.
107. On March 1, 2021, Twitter announced a new five-strike policy as part of the
company’s bar on medical misinformation. COVID-19 misleading information policy, Twitter
(Mar. 1, 2021), https://web.archive.org/web/20210827062904/https://help.twitter.com/en/rules-
and-policies/medical-misinformation-policy. Accounts that committed “[r]epeated violations of
this policy” are subject to increasing levels of discipline, up to and including permanent
suspension for five or more violations. Id.
108. In terms of the mechanics of the five-strike policy, “[t]weets that are labeled and
determined to be harmful will accrue 1 strike.” Id. (emphasis added). Twitter takes no action for
the first strike, but the company imposes a 12-hour-ban for second and third strikes. A fourth
strike yields a one-week suspension with a fifth leading to permanent suspension. Twitter stated
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 38 of 70
39
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
that “[i]ndividuals will be notified directly when a label or required Tweet removal results in
additional account-level enforcement.” Twitter Safety, Updates to our work on COVID-19
vaccine misinformation, Twitter (Mar. 1, 2021),
https://blog.twitter.com/en_us/topics/company/2021/updates-to-our-work-on-covid-19-vaccine-
misinformation. In other words, instead of leaving account holders to guess, the company
committed to giving users notice of the specific tweets violative of the policy.
109. The day after Twitter announced the five-strike policy, Mr. Berenson again
contacted the company, seeking assurances he could continue reporting on COVID-19 using the
platform:
Just following up on this in the wake of Twitter’s announcement yesterday. I intend
to continue to write about the vaccines—as always, with a heavy reliance on
governmental data (whether from Israel, the US, or elsewhere) and published
studies. I appreciate the fact that Twitter has allowed me to provide a risk-benefit
analysis that people are generally not seeing elsewhere and I respect that you do
not want conspiracy theories, etc, on the site. If your fact-checkers do have
questions about something I’ve written, I hope you will let me know and give me a
chance to respond to it before taking any action.
110. Twitter responded the same day, and again the company assured Mr. Berenson
that his work was not being targeted under Twitter’s policies. “I will say that your name has
never come up in the discussions around these policies,” Mr. Borrman said. “If it does I will try
to ensure you’re given a heads up before an action is taken, but I am not always made aware of
them before they’re executed. If something happens, please let me know.
111. With Twitter’s express encouragement, and reasonably believing he had no
reason to fear censorship, Mr. Berenson kept building his Twitter following at the opportunity
cost of time he could have spent on other platforms and activities. He also kept up his relentless,
daily coverage of COVID-19 after Twitter’s launch of the misleading information and five-strike
policy, including the following:
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 39 of 70
40
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
a. On March 25, 2021, Mr. Berenson published his fourth Unreported Truths
booklet, which he promoted on Twitter, and which contained much critical reporting on the
COVID-19 vaccines;
b. He tweeted that “masks are useless” on April 5 and that “civilian masks
are useless” on May 8;
c. On June 24, Mr. Berenson asked whether the CDC “is lying or just
stupid?”; and
d. On July 8, he published the below tweet regarding vaccines and the CDC.
112. Twitter did not label any of these tweets. And since there was no label, the
company did not count any of these tweets as a “strike” against Mr. Berenson’s account under
the five-strike policy.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 40 of 70
41
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
113. Nevertheless, Twitter did label some of Mr. Berenson’s other tweets as
misleading. In a March 15 tweet, Mr. Berenson argued that the vaccines on the traditional
childhood vaccination schedule are “nothing like the mRNA/LNP biotechnology, which is more
properly described as a gene therapy than a vaccine.” Twitter labeled this opinion “misleading.”
As discussed further below, it was not.
114. On May 29 and 30, Twitter labeled four separate tweets as “misleading.” Two of
tweets regarded Chinese-made COVID-19 vaccines. Another presented analysis of COVID-19
data from France and the United Kingdom. The fourth tweet dealt with a spike in COVID-19
deaths following a first vaccine dose. Finally, on July 8, Mr. Berenson tweeted a link to an article
about issues related to transplanting organs from deceased patients “with vaccine-induced
thrombosis and thrombocytopenia” which Twitter labeled “misleading.”
115. Mr. Berenson complained to Mr. Borrman about his company’s labeling of the
March 15 and May 30 tweets. Mr. Borrman stated he would look into the matter, but he did not
respond further. Given Mr. Borrman’s prior assurances and Twitter’s subsequent silence, Mr.
Berenson believed the labeling could have been attributable to the company’s use of algorithms,
which Twitter was still honing.
116. Even with the inconsistent labeling, Mr. Berenson continued to utilize Twitter,
mindful of the company’s prior assurance that his name had “never come up in the discussions
around these policies” as well as Mr. Borrman’s promise to “try to ensure you’re given a heads
up before an action is taken.” Importantly, none of Mr. Berenson’s tweets up until that point had
been labeled as a “strike” under the five-strike policy. At no time during this period did Twitter
lock Mr. Berenson out of his account nor did Twitter take down or delete any of his tweets.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 41 of 70
42
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
117. Under California law, “[a] waiver is an intentional relinquishment of a known
right; may be expressed or implied; [and] any may be implied from conduct inferentially
manifesting an intention to waive. Salton Cmty. Servs. Dist. v. Southard, 256 Cal. App. 2d 526,
532–33, 64 Cal. Rptr. 246, 251 (1967). By knowingly failing to flag these tweets, Twitter waived
any claims that Mr. Berenson violated the five-strike policy based not only on those tweets, but
any future similar statements.
VI. Hours after President Joe Biden says social media companies are “killing people,”
and after other key federal government officials call for social media companies to
censor content, Twitter locks Mr. Berenson’s account for the first time.
118. July 2021 marked the beginning of the end of Mr. Berenson’s relationship with
Twitter. On Saturday, July 10, Mr. Berenson participated in a panel discussion at a political
convention in Dallas, Texas. During the session, Mr. Berenson commented on the government’s
failure to persuade Americans to take one of the available COVID-19 vaccines. The following
day, Dr. Anthony Fauci, President Biden’s Chief Medical Advisor, called Mr. Berenson’s
remarks “horrifying.” According to Dr. Fauci, an official Mr. Berenson had criticized repeatedly
throughout the COVID-19 pandemic, Mr. Berenson was “someone saying that it’s a good thing
for people not to try and save their lives.” Olafimihan Oshin, Fauci: ‘Horrifying’ to hear CPAC
crowd cheering anti-vaccination remarks, The Hill (July 11, 2021),
https://thehill.com/homenews/sunday-talk-shows/562453-fauci-horrifying-to-hear-cpac-crowd-
cheering-missed-vaccine-goal. Far from encouraging people on toward their deaths, Mr.
Berenson was observing that many Americans, particularly younger Americans at demonstrably
less risk of from COVID-19, had simply concluded that the risks of taking one of the COVID-19
vaccines outweighed the benefits.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 42 of 70
43
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
119. The government’s rhetoric intensified as the week went on. A report called “The
Disinformation Dozen,” which identified twelve individuals said to “produce 65% of the shares
of anti-vaccine misinformation on social media platforms,was featured in those efforts.
Shannon Bond, Just 12 People Are Behind Most Vaccine Hoaxes on Social Media, Research
Shows, NPR (May 14, 2021), https://www.npr.org/2021/05/13/996570855/disinformation-dozen-
test-facebooks-twitters-ability-to-curb-vaccine-hoaxes. Many of the people on the list had a
history of general anti-vaccine activism. Mr. Berenson was not included in this blacklist.
120. On July 15, Surgeon General Vivek Murthy published a report titled “Confronting
Health Misinformation.” Vivek H. Murthy, Confronting Health Misinformation: The U.S.
Surgeon General’s Advisory on Building a Healthy Information Environment (2021), available
at https://www.hhs.gov/sites/default/files/surgeon-general-misinformation-advisory.pdf. In the
report, Surgeon General Murthy cited “The Disinformation Dozen” analysis. Id. at 6 n.45. He
noted that “while there have been significant efforts to address health misinformation,” citing
Twitter’s COVID-19 policy, id. at n.49, Surgeon General Murthy said “there is much more to be
done.” He explained “[t]he threat of misinformation raises important questions we must answer
together.” Id. (emphasis in original). In this regard, later in the report, he called for social media
platforms to “[p]rioritize early detection of misinformation ‘super-spreaders’ and repeat
offenders,” and recommended the companies “[i]mpose clear consequences for accounts that
repeatedly violate platform policies.” Id. at 12.
121. In a section of the report captioned “What Governments Can Do,” Surgeon
General Murthy recommended “invest[ing] in fact-checking and rumor control mechanisms
where appropriate.” Id. at 15. As authority for this recommendation, Surgeon General Murthy
cited a report from the Virality Project. See id. at 15 n.62. According to that organization, “[t]he
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 43 of 70
44
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Virality Project supports information exchange between public health officials, government, and
social media platforms through weekly briefings and real-time incident response.Virality
Project Weekly Briefing #31 July 20, 2021 – July 27, 2021, Virality Project, at 1,
https://bit.ly/2YNp2e8. That outlet repeatedly criticized Mr. Berenson’s reporting, at the very
least providing a basis for government officials to become aware of Mr. Berenson and his
reporting if they were not so acquainted previously, if not to identify him as a potential
censorship target. See, e.g., id. at 3 (calling Mr. Berenson “a frequent spreader of vaccine
misinformation”); Virality Project Weekly Briefing #22 May 18, 2021, Virality Project, at 7,
https://bit.ly/2YZufjg (linking to Mr. Berenson’s “prominent anti-vaccine account”).
122. Asked about “actions that the federal government can take to ensure their [(social
media companies’)] cooperation,” Press Secretary Psaki responded that “we are in regular touch
with these social media platforms, and those engagements typically happen through members of
our senior staff.” Id. Press Secretary Psaki revealed that the federal government was “flagging
problematic posts for Facebook that spread disinformation.” Id. Upon information and belief, in
or around July 2021, the federal government also flagged “problematic posts” for Twitter,
including posts by Mr. Berenson.
123. Press Secretary Psaki explained that the federal government had “proposed
changes . . . to social media platforms.” Id. Among them, Ms. Psaki said the government
“recommended—proposed that they [(the companies)] create a robust enforcement strategy that
bridges their properties and provides transparency.” Id. Ms. Psaki even suggested the companies
work together so that people banned on one platform do not “remain active” on others. Id. She
complained that the “disinformation dozen” in particular “remain active on Facebook, despite
some even being banned on other platforms.” Id.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 44 of 70
45
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
124. The following day, July 16, Ms. Psaki again called for social media companies to
lock arms and “bridge their properties” because “[y]ou shouldn’t be banned from one platform
and not others if you—for providing misinformation out there.Press Briefing by Press
Secretary Jen Psaki, July 16, 2021, https://www.whitehouse.gov/briefing-room/press-
briefings/2021/07/16/press-briefing-by-press-secretary-jen-psaki-july-16-2021/. In other words,
from the government’s perspective, censorship on one platform should extend across the entire
social media landscape. Ms. Psaki explained that the government is “in regular touch with social
media platforms . . . about areas where have concern.” Id. Upon information and belief, the
“social media platforms” Ms. Psaki referred to included Twitter.
125. Ms. Psaki further advised that “[a]ny decision about platform usage and who
should be on the platform is orchestrated and determined by private-sector companies.” Id. That
is “their decision to do,” Ms. Psaki explained. Id. “That is not the federal government doing
that.” Id. But echoing Dr. Fauci’s statements five days earlier, she said “[i]t is life and death. It is
a public health issue in the country.Id.
126. Picking up on these themes, at around the same time of the White House press
briefing, in response to a reporter who asked, “On COVID misinformation, what’s your message
to platforms like Facebook,” President Biden said, “They’re killing people.” President Biden:
“They’re killing people”, C-SPAN, July 16, 2021,
https://www.youtube.com/watch?v=gJoOtLn4goY. President Biden’s statement caused other
media to conclude that the government “blamed” social media companies “for spreading
misinformation about the coronavirus and vaccines,” creating “stalling U.S. vaccine rates.”
Lauren Egan, ‘They’re killing people’: Biden blames Facebook, other social media for allowing
Covid misinformation, NBC News (July 16, 2021, 4:10 PM EDT),
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 45 of 70
46
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
https://www.nbcnews.com/politics/white-house/they-re-killing-people-biden-blames-facebook-
other-social-media-n1274232. Tellingly and, upon information and belief, not coincidentally,
Mr. Berenson’s comments which Dr. Fauci attacked as “horrific” concerned the same subject—
vaccine rates that had plateaued despite the government’s efforts.
127. Less than four hours after President Biden’s comment and Press Secretary Psaki’s
briefing, Twitter locked Mr. Berenson’s account for the first time. Under the five-strike policy,
the lock was apparently his second strike, though the company did not inform him of the first
strike. Despite prior assurances that the company would try to give Mr. Berenson a “heads up”
before taking action, Twitter gave him no advance notice prior to locking his account. Upon
information and belief, Twitter did not make any effort to notify Mr. Berenson.
128. As for its part, the government kept pressing in. Four days after President Biden’s
comments, reports circulated that “[t]he White House is assessing whether social media
platforms are legally liable for misinformation spread on their platforms.” Matthew Brown,
‘They should be held accountable’: White House reviews platforms’ misinformation liability,
USA Today, July 20, 2021, https://www.usatoday.com/story/news/politics/2021/07/20/white-
house-reviews-section-230-protections-covid-misinformation/8024210002/. The report noted
“[r]elations are tense between the Biden administration and social media platforms,” and that the
government was “examining how misinformation fits into the liability protections granted by
Section 230 of the Communications Decency Act, which shields online platforms from being
responsible for what is posted by third parties on their sites.” Id.
129. Upon information and belief, in July 2021, the government encouraged Twitter to
censor Mr. Berenson’s account. Twitter says that it is “enforcing this policy in close
coordination with trusted partners, including public health authorities, NGOs and governments,
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 46 of 70
47
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
and continue to use and consult with information from those sources when reviewing content.”
COVID-19 misleading information policy, Twitter (Dec. 16, 2020),
https://web.archive.org/web/20201216200114/https://help.twitter.com/en/rules-and-
policies/medical-misinformation-policy (emphasis added). Moreover, the government’s threats
to take away section 230 protection on account of social media companies’ failure to take
sufficient steps from the government’s perspective to combat medical misinformation provided
additional pressure to censor.
130. On September 7, Senator Elizabeth Warren appeared to commend Twitter for its
COVID-19 misinformation policy, and its decision to ban Mr. Berenson. In a letter to Amazon,
Senator Warren suggested that the bookseller apply a similar standard to COVID-19 information
in books and other materials available on that platform. Letter from Senator Elizabeth Warren to
Andy Jassy, Chief Executive Officer, Amazon, Inc., Sept. 7, 2021, at 5,
https://www.warren.senate.gov/imo/media/doc/2021.9.7%20Letter%20to%20Amazon%20on%2
0COVID%20Misinformation.pdf. In this regard, Senator Warren noted that Twitter “recently
ban[ned] Alex Berenson—whose books appear prominently in Amazon searches—from its
platform.” Id.
131. “The Supreme Court has articulated four tests for determining whether a private
individual’s actions amount to state action: (1) the public function test; (2) the joint action test;
(3) the state compulsion test; and (4) the governmental nexus test.” Franklin v. Fox, 312 F.3d
423, 444-45 (9th Cir. 2002). Twitter’s censorship of Mr. Berenson constitutes state action under
at least the joint action and governmental nexus tests.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 47 of 70
48
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
VII. Twitter permanently suspends Mr. Berenson, violating its rules and breaking its
prior promises in the process.
132. Twitter’s Terms of Service (the “Terms”) are one of several documents that
purport to govern the legal relationship between the company and account holders. Twitter
Terms of Service, https://twitter.com/en/tos (updated August 19, 2021).
3
Individual account
owners have no ability to comment, revise, or even to propose revisions to the Terms. The Terms
are a contract of adhesion. California law requires such contracts to be construed against the
drafter, which in this case is Twitter. See City of Hope Nat’l Med. Ctr. v. Genentech, Inc., 43 Cal.
4th 375, 397, 181 P.3d 142, 158 (2008) (referring to the “special construction rules” that apply to
adhesion contracts such as construing terms against the drafter).
133. The Terms provide that the company “may suspend or terminate your account 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 and Policies.” Id. As described further below, Twitter did not cite this passage to justify
banning Mr. Berenson from its platform.
134. Also relevant here, the Terms contain no merger or integration clause. The Terms
contain no bar on modification by course of conduct or from trade usage. Under California law,
“[t]he terms may be explained or supplemented by course of dealing or usage of trade or by
course of performance.” Cal. Civ. Proc. Code § 1856(c).
135. Twitter gave Mr. Berenson no personal notice of his suspension or any
explanation as to why he was suspended. Instead, Mr. Berenson learned through media reports
3
Twitter modified the Terms on June 18, 2020 and January 1, 2020, but those modifications
have no bearing on this dispute.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 48 of 70
49
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
that Twitter’s permanent suspension of his account was not based on any violation of the Terms.
As the company informed NBC, Fox News, and other media outlets, Mr. Berenson was
“permanently suspended for repeated violations of our COVID-19 misinformation rules.
Michael Ruiz, Twitter permanently suspends Alex Berenson over coronavirus tweets, Fox
Business, Aug. 28, 2021, https://www.foxbusiness.com/media/twitter-permanently-suspends-
alex-berenson-over-coronavirus-tweets.
136. Twitter’s COVID-19 misinformation rules provide a definite, objective standard.
The operative medical information policy provides that “for content related to COVID-19 to be
labeled or removed under this policy,” which is also the predicate for an account incurring a
“strike,” the content “must” do the following: (1) “advance a claim of fact, expressed in
definitive terms; (2) “be demonstrably false or misleading, based on widely available,
authoritative sources”; and (3) “be likely to impact public safety or cause serious harm.” The
five-strike system itself uses the phrase “we will” to describe the policy. The policy contains no
reservation of rights to permanently suspend an account or bar a user from the platform for “any
or no reason” as the Terms purport to do. Further, the policy was issued after the operative
Terms governing Mr. Berenson and Twitter’s relationship. To suspend Mr. Berenson from the
platform on the basis that he violated these rules, then he must have objectively violated the
rules. Mr. Berenson did not do so.
137. To this day, Mr. Berenson is unaware which tweet was his first strike. His second
strike, which coincided with President Biden’s charge that social media companies were “killing
people,” was for the following tweet:
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 49 of 70
50
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
In the tweet, Mr. Berenson linked to another user’s analysis of COVID-19 case counts in
countries with high vaccination rates
138. The tweet did not violate Twitter’s policy, for at least five reasons. First, Twitter
never labeled the underlying, linked-to tweet as “misleading,” the only factual content in Mr.
Berenson’s tweet.
4
Second, two days prior to strike two, Twitter took no action when Mr.
Berenson tweeted his opinion that “the vaccines are failing,” which was based on data that “[t]he
UK, the most vaccinated major country in the world, reported 42,000 cases” of COVID-19, a
“100x” increase from the previous year. Third, Mr. Berenson tweeted his opinion, and the policy
expressly provides for protection of opinion. Fourth, Mr. Berenson’s reporting in this tweet was
much like the reporting Twitter was previously aware of and did not object to. Fifth, despite the
company’s previous assurances that it would try to give Mr. Berenson a “heads up” prior to
acting, Twitter gave him no notice prior to applying the strike.
4
To this day, the underlying tweet is still not labeled. See Corona Realism (@holmenkollin),
Twitter (July 16, 2021, 6:59 AM),
https://twitter.com/holmenkollin/status/1415989536933490688.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 50 of 70
51
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
139. On July 27, Twitter locked Mr. Berenson’s account for twelve hours again, his
third strike on the platform. The company did not identify the tweet that allegedly violated the
Twitter policy. Again, Twitter did not give Mr. Berenson notice prior to locking his account.
140. Twitter gave Mr. Berenson his fourth strike on July 30 based on the following:
141. This tweet was true, and not false or misleading. Mr. Berenson cited a pre-print,
non-peer-reviewed review of six month safety efficacy data for the Pfizer-BioNTech COVID-19
vaccine. Stephen J. Thomas et al., Six Month Safety and Efficacy of the BNT162b2 mRNA
COVID-19 Vaccine, July 28, 2021,
https://www.medrxiv.org/content/10.1101/2021.07.28.21261159v1.full.pdf. The report noted
that “[d]uring the blinded, controlled period, 15 BNT162b2 and 14 placebo recipients died.” Id.
at 6. With this, Mr. Berenson was pointing out that this data showed there was no significant
difference in overall mortality between patients who received the Pfizer-BioNTech COVID-19
vaccine and the placebo. To be clear, his comment was not about whether the vaccine studied
prevented patients from contracting COVID-19 or whether patients who took the vaccine had
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 51 of 70
52
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
less severe cases of the disease. It was an evidence-based interpretation of overall mortality data
and nothing more than that.
142. Mr. Berenson’s reference to the “trial blind” referred to the fact that Pfizer gave
patients in the placebo arm of the clinical trial a COVID-19 vaccine after the product received
emergency use authorization. As a result, it would not be possible to track long-term overall
mortality between the two groups from the original trial without a comparator placebo group.
This is what Mr. Berenson meant when he tweeted that the “trial blind is broken now.”
143. Critiquing or otherwise questioning clinical trial design, particularly the
circumstances around breaking trial blinds, could hardly be considered controversial or
damaging to the public. In a 1998 guidance document, FDA noted that “[b]reaking the blind (for
a single subject) should be considered only when knowledge of the treatment assignment is
deemed essential by the subject’s physician for the subject’s care.” FDA, Guidance for Industry,
E9 Statistical Principles for Clinical Trials, at 12 1998), available at
https://www.fda.gov/media/71336/download.
144. A month later, during the evening of August 28, Twitter issued its fifth strike
against Mr. Berenson, permanently suspending his account. The company did not cite the
specific tweet that formed the basis of the fifth strike. Upon information and belief, the tweet
below, which the company labeled misleading, served as Twitter’s basis for the fifth strike and
Mr. Berenson’s permanent suspension from the platform.
///
///
///
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 52 of 70
53
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
145. Like the other strikes, this tweet also did not violate Twitter’s stated policy. Mr.
Berenson’s claim that the COVID-19 vaccines do not “stop infection” or “transmission” of
COVID-19 was true at the time and is true now. It is undisputed that vaccinated persons can
contract and spread COVID-19. On July 30, Dr. Anthony Fauci confirmed that “individuals who
are infected through breakthrough infections, namely vaccinated people who ultimately get
infected, that they are generally without symptoms or minimally symptomatic, however it is
clear that they are capable of transmitting the infection to uninfected individuals.” Dr. Fauci on
COVID-19 Spread, Yahoo Finance, July 31, 2021, https://youtu.be/mP9iHyj1uiU.
146. Dr. Fauci’s July 30 observation remains true today. In a December 2021 editorial
in The New England Journal of Medicine, Dr. Fauci and two other medical doctors explained
that “[a]s important as [the COVID-19 vaccines] are, however, their protective efficacy wanes
over time, necessitating booster doses,” and that “[v]accination has also been unable to prevent
‘breakthrough’ infections, allowing subsequent transmission to other people even when the
vaccine prevents severe and fatal disease.” David M. Morens, Jeffery K. Taubenberger, M.D.,
Ph.D., and Anthony S. Fauci, Universal Coronavirus Vaccines—An Urgent Need, New Engl. J.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 53 of 70
54
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Med., Dec. 15, 2021, https://www.nejm.org/doi/full/10.1056/NEJMp2118468. If anything, this
article corroborates Mr. Berenson’s allegedly ban-worthy tweet, rather than contradict it.
147. What is more, Twitter amended its COVID-19 misinformation policy in
December 2021. On or around December 10, Twitter announced that it would label claims “that
people who received the vaccine can spread or shed the virus (or symptoms or immunity) to
unvaccinated people.” Twitter, COVID-19 misleading information policy, Dec. 10, 2021,
https://web.archive.org/web/20211210152757/https://help.twitter.com/en/rules-and-
policies/medical-misinformation-policy. That change, made more than three months after Mr.
Berenson’s August 28 tweet, would have arguably brought his prior statement within the scope
of the policy. But then Twitter reversed course. On December 16, the day after The New England
of Journal Medicine article was published, Twitter revised its policy to say it would label claims
that “people who have received the vaccine can spread or shed the vaccine (or symptoms, or
immunity)” as opposed to the virus. Twitter, COVID-19 misleading information policy, Dec. 16,
2021, https://web.archive.org/web/20211216085837/https://help.twitter.com/en/rules-and-
policies/medical-misinformation-policy. What that means is that, even today, Mr. Berenson’s
allegedly ban-worthy claim that the COVID-19 vaccines do not stop infection or transmission
does not violate Twitter’s policy.
148. Snopes reported that “[a]ccording to the company, the use of the word ‘virus’ in
that context,” i.e., in the December 10 iteration of the policy, “was merely a typo, which was
corrected on Dec. 16, and did not reflect the true policy anyway.” Dan MacGuill, Did Twitter
Prohibit Making an Accurate Claim About COVID-19 Transmission?, Snopes, Dec. 16, 2021,
https://www.snopes.com/fact-check/twitter-ban-covid-vaccine/. The Snopes article quotes a
Twitter spokesperson as follows:
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 54 of 70
55
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
The earlier language is not reflective of our enforcement approach, past or present.
We’ve enforced our COVID-19 misleading information policy in line with the
corrected language only. We do not take enforcement action against claims that
people who receive the vaccine can spread or shed the virus.
Id. (emphasis added). And yet, in Mr. Berenson’s case, contrary to this claim, Twitter took action
against his August 28 tweet, including his assertion that the vaccines do not stop transmission of
COVID-19.
149. With respect to the rest of the tweet, Mr. Berenson’s recommendation not to think
of the COVID-19 vaccines “as a vaccine” turns on the definition of that term. Even though the
company could have done so, Twitter’s policy does not define the term, so outside sources must
be consulted. In a glossary on its website, the CDC defines the term in relevant part as a
“suspension of live (usually attenuated) or inactivated microorganisms (e.g. bacteria or viruses)
or fractions thereof administered to induce immunity and prevent infectious diseases and their
sequelae.” CDC, Vaccines & Immunizations, Glossary,
https://www.cdc.gov/vaccines/terms/glossary.html (last updated July 30, 2020) (emphasis
added). Similarly, FDA explains on its website that “[v]accines work by mimicking the
infectious bacteria or viruses that cause disease. Vaccination stimulates the body’s immune
system to build up defenses against the infectious bacteria or virus (organism) without causing
the disease.” FDA, Vaccine Development – 101, Dec. 14, 2020, https://www.fda.gov/vaccines-
blood-biologics/development-approval-process-cber/vaccine-development-101 (emphasis
added). Courts have similarly defined the term. See, e.g., Dean v. Sec’y of Health & Hum. Servs.,
No. 16-1245V, 2018 WL 3104388, at *9 (Fed. Cl. May 29, 2018) (defining “vaccine” to be “any
substance designed to be administered to a human being for the prevention of 1 or more
diseases”) (quoting 26 U.S.C. § 4132(a)(2)). Finally, California’s Sherman Food, Drug, and
Cosmetic Law contains two references to the term “vaccine,” with the one relevant reference
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 55 of 70
56
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
relating the term to preventing infection. See Cal. Health & Safety Code § 111605 (defining
“AIDS-related drug” to include a “vaccine to protect against human immunodeficiency virus
(HIV) infection”).
150. Set against this backdrop, it was not false or misleading for Mr. Berenson to claim
that the available COVID-19 vaccines are not “vaccines” in the traditional sense, but rather
“therapeutics with a limited efficacy window.” Further, Mr. Berenson’s characterizations of the
side effect profile of the vaccines were at very least protected opinions under the terms of
Twitter’s policy as were his critiques of vaccine mandates.
151. Twitter gave no notice to Mr. Berenson prior to issuing its fifth strike and
permanent suspension from the platform. Again, this was contrary to the prior representations
made to him by a company executive. The company also denied Mr. Berenson’s appeal to be
reinstated to the platform. Further, the tweets that precipitated his suspension from Twitter, the
five strikes, were all substantially similar in content and tone to tweets he made in the past with
Twitter’s knowledge, consent, and apparent approval. Again, Twitter executive Brandon
Borrman assured Mr. Berenson, even after he started to tweet about the vaccines, that “your
name has never come up in the discussions around these policies.”
152. Recall that to violate Twitter’s COVID-19 misinformation policy, a tweet “must,”
in addition to being “demonstrably false or misleading, based on widely available, authoritative
sources,” also “be likely to impact public safety or cause serious harm.” Upon information and
belief, Twitter has no direct evidence that any of the purported five strikes against Mr.
Berenson’s account affected public safety or caused serious harm.
153. Mr. Berenson did not violate Twitter’s COVID-19 rules. As such, requiring
Twitter to carry his messages does not raise compelled speech problems under the First
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 56 of 70
57
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Amendment, because the company would not be required to broadcast messages with which it
disagrees—i.e., messages that violate the company’s rules.
154. In the months leading up to his suspension from the platform, Mr. Berenson’s
Twitter following continued to grow. Just before he was suspended, Mr. Berenson claimed
345,000 followers, a nearly fifty-fold increase over his following from March 2020. In the two
months prior to his suspension from Twitter, his tweets received 180 and 182 million views,
respectively, while his Twitter profile was viewed 19.5 and 25 million times during the same
time period. Upon information and belief, at all relevant times, to include during July and August
2021, Twitter profited from Mr. Berenson’s reporting on the platform.
155. Twitter itself competes against Mr. Berenson for an audience regarding COVID-
19. Twitter also profits from use of its platform by other media outlets, to include outlets and
journalists that compete against Mr. Berenson. With respect to COVID-19, the company
acknowledges that “[m]ore people than ever are coming to Twitter to learn about COVID-19.”
To capitalize on this audience, Twitter explained it is rolling out “experimental settings [to] give
newsrooms and journalists over the conversations they start.” Twitter, COVID-19 news coverage
on Twitter, https://media.twitter.com/en/articles/best-practice/2020/covid-19-news-coverage-on-
twitter (last visited Dec. 19, 2021). The page itself features NBC, Reuters, CBS, and CNN
among others.
156. Twitter’s branding of Mr. Berenson as a spreader of “misinformation” along with
the company’s complete censorship of his speech provides itself and media outlets the company
services with a competitive advantage over Mr. Berenson. Twitter profits not only from the
traffic generated by these competitive outlets, but from any ancillary services adjacent to its
platform it might be able to sell. At the same time, Twitter’s censorship of Mr. Berenson garners
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 57 of 70
58
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
the company favor from powerful governmental actors, including the current Administration and
members of Congress.
CLAIMS FOR RELIEF
COUNT I
VIOLATION OF THE FIRST AMENDMENT
Free Speech Clause
Petition Clause
U.S. Const. Amend. I
157. Mr. Berenson re-alleges the preceding paragraphs herein.
158. In Bivens v. Six Unknown Named Agents, 403 U.S. 388 (1971), the Supreme
Court created an implied cause of action for violations of the Fourth Amendment.
159. The Ninth Circuit has recognized the applicability of Bivens to certain First
Amendment claims. See Gibson v. United States, 781 F.2d 1334, 1342 (9th Cir. 1986) (“We now
follow these circuits in holding that, because plaintiffs have alleged that FBI agents acted with
the impermissible motive of curbing . . . protected speech, they have asserted a claim properly
cognizable through a Bivens-type action directly under the First Amendment.”).
160. Twitter deprived Mr. Berenson of the right to speak on its platform.
161. At the same time, Twitter deprived Mr. Berenson of the right to petition the
federal government for a redress of grievances by tweeting to and interacting with accounts run
and managed by the federal government, including @POTUS.
162. As outlined in detail above, Twitter deprived Mr. Berenson of these rights as part
of its “partnership” with the federal government. Upon information and belief, the federal
government enticed and coerced Twitter to deprive Mr. Berenson of his right to speak and
petition the federal government on the platform.
163. Twitter’s actions in this case constitute state action under the U.S. Constitution.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 58 of 70
59
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
164. Twitter’s wholesale censorship of Mr. Berenson’s speech and petition rights
violates the First Amendment and is actionable as a Bivens claim.
COUNT II
FEDERAL FALSE ADVERTISING AND UNFAIR COMPETITION
Lanham Act § 43(a)(1)(B), 15 U.S.C. § 1125(a)(1)(B)
165. Mr. Berenson re-alleges the preceding paragraphs herein.
166. Twitter made false and misleading representations concerning Mr. Berenson’s
professional activities in commerce, namely, his reporting on COVID-19, including without
limitation the following:
a. That Mr. Berenson committed “repeated violations” of Twitter’s COVID-
19 misinformation policies;
b. That Mr. Berenson’s tweets regarding Chinese-made COVID-19 vaccines
were misleading;
c. That Mr. Berenson’s tweet regarding the Pfizer-BioNTech COVID-19
vaccine trial was false or misleading;
d. That a tweet from Mr. Berenson regarding an increase in COVID-19
deaths following a first vaccine dose was misleading; and
e. That Mr. Berenson’s tweet linking to an article about issues related to
transplanting organs from deceased patients “with vaccine-induced thrombosis and
thrombocytopenia” was misleading.
167. Twitter’s false and misleading representations regarding Mr. Berenson have
actually deceived and are likely to deceive consumers regarding Mr. Berenson’s reporting
services and cause them to refrain from reading Mr. Berenson’s reporting and/or purchasing his
books or pamphlets. At the same time, Twitter’s actions make it more likely these consumers get
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 59 of 70
60
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
news from other sources approved by Twitter, to the company’s direct and indirect economic
benefit.
168. Twitter’s false and misleading representations regarding Mr. Berenson were
disseminated in interstate commerce and were made in connection with commercial advertising
and/or promotion of Twitter as a source for information concerning COVID-19.
169. Mr. Berenson’s interests in this case fall within the zone of interests protected by
the Lanham Act. Twitter’s false and misleading representations regarding Mr. Berenson harm his
ability to retain and expand his readership and audience, including the relevant market for his
reporting and for his books and pamphlets.
170. These harms suffered by Mr. Berenson are the proximate result of Twitter’s false
and misleading representations.
171. This claim concerns Twitter’s own statements. Even so, “immunity under [section
230] does not extend to anticompetitive conduct.” Enigma Software, 946 F.3d at 1054. To the
extent Twitter’s false and misleading representations conferred a competitive advantage on itself
and Mr. Berenson’s journalistic competitors, the company cannot claim the benefit of section
230 immunity.
COUNT III
VIOLATION OF CALIFORNIA COMMON CARRIER LAW
Cal. Civ. Code §§ 2168, 2169, 2209
172. Mr. Berenson re-alleges the preceding paragraphs herein.
173. Under California law, “[e]very one who offers to the public to carry persons,
property, or messages, excepting only telegraphic messages, is a common carrier of whatever he
thus offers to carry.” Cal. Civ. Code § 2168. In California, “a common carrier must, if able to do
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 60 of 70
61
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
so, accept and carry whatever is offered to him, at a reasonable time and place, of a kind that he
undertakes or is accustomed to carry.” Id. § 2169.
174. Further, “[e]very person whose message is refused or postponed, contrary to the
provisions of this Chapter, is entitled to recover from the carrier his actual damages, and fifty
dollars in addition thereto.Id. § 2209.
175. Twitter offers to the public to carry “messages” on its platform.
176. Twitter is not a telegraphic message service.
177. Twitter’s Terms provide that “[t]he of the State of California, excluding its choice
of law provisions, will govern . . . any dispute that arises” between a user and the company.
178. Twitter has refused and currently refuses to carry Mr. Berenson’s messages. The
company denied his appeal to be reinstated to the platform.
179. Mr. Berenson has incurred damages by Twitter’s refusal, including but not limited
to on account of not being to publish his reporting, promote his booklets and new book, and sell
subscriptions to his Substack account.
180. Twitter’s refusal to carry Mr. Berenson’s messages violates at least sections 2169
and 2209 of the California Civil Code.
181. California’s common carrier law protects speech against private censorship.
182. To the extent section 230 applies to the California common carrier law, a federal
statute “is the source of the power and authority by which any private rights are lost or
sacrificed.” Hanson, 351 U.S. at 232.
183. Section 230 is then “the governmental action on which the Constitution operates,
though it takes a private agreement,” or in the case of social media companies, private actions,
“to invoke the federal sanction.Id.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 61 of 70
62
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
184. Using section 230, a federal statute, to nullify the California common carrier
statute to facilitate private censorship of Mr. Berenson’s speech violates the First Amendment.
185. Requiring Twitter to carry Mr. Berenson’s messages would not violate the
company’s First Amendment rights. Mr. Berenson did not violate Twitter’s Terms or its rules.
As a result, application of section 2168 to Twitter would not require the company to carry speech
that violates its policies.
COUNT IV
VIOLATION OF CALIFORNIA UNFAIR COMPETITION LAW
Cal. Bus. & Professions Code §§ 17200 et seq.
186. Mr. Berenson re-alleges the preceding paragraphs herein.
187. The California Unfair Competition Law (the “UCL”) prohibits “any unlawful,
unfair or fraudulent business act or practice and unfair, deceptive, untrue or misleading
advertising and any act prohibited by [California’s False Advertising Law].” Cal. Bus. & Prof.
Code § 17200.
188. At a minimum, the following actions by Twitter against Mr. Berenson were
unfair, deceptive, and untrue with respect to Mr. Berenson:
a. The statement that Mr. Berenson committed “repeated violations” of
Twitter’s COVID-19 misinformation policies;
b. Labeling Mr. Berenson’s tweets regarding Chinese-made COVID-19
vaccines as misleading;
c. Labeling Mr. Berenson’s tweet regarding the Pfizer-BioNTech COVID-19
vaccine trial as false or misleading;
d. Labeling Mr. Berenson tweet regarding an increase in COVID-19 deaths
following a first vaccine dose as misleading; and
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 62 of 70
63
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
e. Labeling Mr. Berenson’s tweet linking to an article about issues related to
transplanting organs from deceased patients “with vaccine-induced thrombosis and
thrombocytopenia” as misleading.
189. Further, to the extent Twitter maintains that the COVID-19 misinformation rules
are not to be objectively read and applied, and that the company can label content regardless of
what its rules say, then such is also unfair, deceptive, and untrue.
190. “An unlawful business practice under [UCL] section 17200 is ‘an act or practice,
committed pursuant to business activity, that is at the same time forbidden by law.’” Hale v.
Sharp Healthcare, 183 Cal. App. 4th 1373, 1383, 108 Cal. Rptr. 3d 669, 677 (2010) (citing
Progressive West Ins. Co. v. Superior Court, 135 Cal. App. 4th 263, 287, 37 Cal. Rptr. 3d 434,
453 (2005)). Provided the conduct at issue “is at the time forbidden by law,” then “[v]irtually
any law—federal, state or local—can serve as a predicate for an action under” the UCL. Id.
191. To have standing under the UCL, a plaintiff must “(1) establish a loss or
deprivation of money or property sufficient to qualify as injury in fact, i.e., economic injury, and
(2) show that that economic injury was the result of, i.e., caused by, the unfair business practice
or false advertising that is the gravamen of the claim.” Kwikset Corp. v. Superior Ct., 51 Cal. 4th
310, 322, 246 P.3d 877, 885 (2011).
192. At a minimum, Twitter’s refusal to carry Mr. Berenson’s messages constitutes a
violation of the California common carrier law. Mr. Berenson has suffered a specific economic
injury on account of Twitter’s violation of that statute, including but not limited to an account of
not being able to publish his reporting, promote his booklets and new book, and sell
subscriptions to his Substack account.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 63 of 70
64
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
193. To the extent section 230 applies to the UCL, a federal statute “is the source of
the power and authority by which any private rights are lost or sacrificed.” Hanson, 351 U.S. at
232.
194. Section 230 is then “the governmental action on which the Constitution operates,
though it takes a private agreement,” or in the case of social media companies, private actions,
“to invoke the federal sanction.Id.
195. Using section 230, a federal statute, to nullify the UCL to facilitate private
censorship of Mr. Berenson’s speech violates the First Amendment.
196. Moreover, “immunity under [section 230] does not extend to anticompetitive
conduct.” Enigma Software, 946 F.3d at 1054. To the extent Twitter’s censorship conferred a
competitive advantage on itself and Mr. Berenson’s journalistic competitors, the company cannot
claim the benefit of section 230 immunity.
197. Further, though section 230(e)(2), which provides that “[n]othing in this section
shall be construed to limit or expand any law pertaining to intellectual property,” has been read
to apply only to limited federal intellectual property claims in the Ninth Circuit, Perfect 10, Inc.
v. CCBill LLC, 488 F.3d 1102 (9th Cir. 2007), the Third Circuit recently held that the provision
“applies to state intellectual property law,” Hepp v. Facebook, 14 F.4th 404, 212 (3d Cir. 2021).
Hepp reflects the Supreme Court’s commitment to textualism, see Bostock v. Clayton Cty.,
Georgia, 140 S. Ct. 1731, 1737 (2020) (“Only the written word is law, and all persons are
entitled to its benefits.”), and should be applied here to find Mr. Berenson’s UCL outside the
scope of section 230, because unfair competition is a subspecies of intellectual property law.
COUNT V
BREACH OF CONTRACT
198. Mr. Berenson re-alleges the preceding paragraphs herein.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 64 of 70
65
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
199. The Terms and Twitter’s rules, to include the COVID-19 medical information
policy, constitute a binding contract between Twitter and Mr. Berenson.
200. The contract between Twitter and Mr. Berenson is a contract of adhesion under
California law.
201. Since Twitter is the drafter of the terms of the contract between it and Mr.
Berenson, the terms should be construed against the company.
202. Twitter amended its Terms by promulgating its COVID-19 five-strike policy.
203. Twitter’s rules committed the company to a specific five-strike policy under
which content must meet certain defined, objective criteria to be deemed violative. The company
is bound to apply that policy objectively and rationally under California’s duty of good faith and
fair dealing which requires, among other things, “each contracting party to refrain from doing
anything to injure the right of the other to receive the benefits of the agreement.Egan v. Mut. of
Omaha Ins. Co., 24 Cal. 3d 809, 818, 620 P.2d 141 (1979).
204. As set forth above, the content Mr. Berenson published did not violate any of
Twitter’s rules, as defined.
205. Twitter breached its contract with Mr. Berenson by permanently suspending his
account on the basis that he violated the company’s COVID-19 misinformation rules.
206. Mr. Berenson incurred damages on account of Twitter’s breach of the contract.
207. Twitter’s breach of contract was intentional and willful.
208. Since Twitter is a common carrier and its actions against Mr. Berenson were at
least willful, the liability waiver the company unilaterally included in its Terms is unenforceable
since a “common carrier cannot be exonerated, by any agreement made in anticipation thereof,
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 65 of 70
66
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
from liability for the gross negligence, fraud, or willful wrong of himself or his servants.” Cal.
Civ. Code § 2175.
COUNT VI
PROMISSORY ESTOPPEL
209. Mr. Berenson re-alleges the preceding paragraphs herein.
210. Twitter made unambiguous, personal promises and assurances to Mr. Berenson
over the course of more than a year, including but not limited to the following:
a. That Mr. Berenson was raising “nuanced points” about the COVID-19
pandemic and the response to the same;
b. That Twitter was taking a “nuanced approach” to discussion around
COVID-19;
c. That Twitter was “trying to make sure that factual debate finds a way
through the emotion”;
d. That Twitter was “trying to avoid appearing as though we’re saying one
treatment or solution [to COVID-19] is good or bad”;
e. That Twitter is “leading with labeling here, rather than simply deciding
between leave up/take down”;
f. That Twitter did not “expect that you will see major change in how we are
addressing things going forward”;
g. That the points Mr. Berenson was raising about COVID-19 vaccines in
and around December 2020 “should not be an issue at all”;
h. That, as of March 2021, Mr. Berenson’s “name has never come up in the
discussions around these policies”; and
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 66 of 70
67
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
i. That Twitter would “try to ensure you’re given a heads up before an action
is taken” against Mr. Berenson’s account.
211. These specific, personal promises amplified the general promises contained in the
five-strike policy under which Twitter purported to permanently ban Mr. Berenson. California
courts have held vaguer and more open-ended statements than those Twitter made to Mr.
Berenson sufficiently definite to support a promissory estoppel claim. See Garcia v. World Sav.,
FSB, 183 Cal. App. 4th 1031, 1045–46, 107 Cal. Rptr. 3d 683, 696 (2010) (surveying cases
where courts found various statements to be sufficiently definite, including a “promise to use
‘best efforts to acquire land sufficient to support promissory estoppel” and “borrowers promise
that ‘if’ he obtained earthquake insurance, lender would be loss payee”).
212. Further, Twitter had specific knowledge of Mr. Berenson’s tweets, and did not
object to them. Instead, the company enticed Mr. Berenson to continue using its service.
213. Mr. Berenson relied on Twitter’s promises and assurances by continuing to use
the platform.
214. Mr. Berenson’s reliance was reasonable and foreseeable.
215. Mr. Berenson was injured by his reliance by continuing to build his following on
Twitter as opposed to investing time and attention on other available platforms such as Telegram
and Substack.
COUNT VII
VIOLATION OF CALIFORNIA CONSTITUTION
Free Speech Clause
Petition Clause
Cal. Const. art. I, §§ 2, 3
216. Mr. Berenson re-alleges the preceding paragraphs herein.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 67 of 70
68
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
217. The California Supreme Court has interpreted “sections 2 and 3 of article I of the
California Constitution protect speech and petitioning, reasonably exercised, in shopping centers
even when the centers are privately owned.” Robins v. Pruneyard Shopping Ctr., 23 Cal. 3d 899,
910, 592 P.2d 341, 347 (1979).
218. Like the shopping center in PruneYard, Twitter is held open to the public.
Further, unlike the shopping center in that case, Twitter’s express purpose is to facilitate speech
and communication about issues.
219. The rule in Pruneyard has not yet been extended to social media companies, but
the logic of the case warrants application to Twitter.
220. Under Pruneyard, and if extended, Twitter’s censorship of Mr. Berenson’s speech
violates the California State Constitution.
221. To the extent section 230 applies to the California Constitution, a federal statute
“is the source of the power and authority by which any private rights are lost or sacrificed.”
Hanson, 351 U.S. at 232.
222. Section 230 is then “the governmental action on which the Constitution operates,
though it takes a private agreement,” or in the case of social media companies, private actions,
“to invoke the federal sanction.Id.
223. Using section 230, a federal statute, to nullify the California Constitution to
facilitate private censorship of Mr. Berenson’s speech violates the First Amendment.
COUNT VIII
UNJUST ENRICHMENT
224. Mr. Berenson re-alleges the preceding paragraphs herein.
225. Twitter claims that it banned Mr. Berenson for repeatedly violating the platform’s
medical misinformation rules.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 68 of 70
69
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
226. According to Twitter, for at least a certain time period, Mr. Berenson used Twitter
in violation of the company’s policies, and Twitter profited in the process.
227. Any profits traceable to Mr. Berenson’s reporting are a benefit conferred on
Twitter that it should not have received.
228. Mr. Berenson is entitled to restitution from Twitter.
PRAYER FOR RELIEF
WHEREFORE, Mr. Berenson respectfully prays that this Court:
A. Enter an injunction requiring Twitter to comply with California’s common carrier
statute, the UCL, its contractual obligations, and the promises the company made to Mr.
Berenson by requiring Twitter to carry Mr. Berenson’s messages as it does other members of the
public;
B. Enjoin Twitter from violating Mr. Berenson’s rights under the federal Lanham
Act and the California and U.S. Constitutions;
C. Award Mr. Berenson damages under the federal Lanham Act, the California
common carrier law, and the UCL;
D. Award Mr. Berenson damages for breach of contract;
E. Award Mr. Berenson reliance damages for promissory estoppel;
F. Disgorge Twitter of its ill-gotten gains attributable to Mr. Berenson’s use of
Twitter; and
G. Award any other relief to Mr. Berenson which this Court deems necessary and
proper.
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 69 of 70
70
COMPLAINT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
DEMAND FOR TRIAL BY JURY
Mr. Berenson requests a jury trial on all issues so triable.
Respectfully submitted, this 20th day of December 2021.
ENVISAGE LAW
By: /s/ James R. Lawrence, III
James R. Lawrence, III*
Anthony J. Biller*
* pro hac vice pending
CHARIS LEX P.C.
By: /s/ Sean P. Gates
Sean P. Gates
Attorneys for Plaintiff Alex Berenson
Case 3:21-cv-09818 Document 1 Filed 12/20/21 Page 70 of 70