Penske Media Corporation v. Google LLC

UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA _________________________________________ ) CHEGG, INC., ) ) Plaintiffs, ) ) v. ) Case No. 25-cv-00543 (APM) ) GOOGLE LLC, et al., ) ) Defendants. ) _________________________________________ ) _________________________________________ ) PENSKE MEDIA CORPORATION, et al., ) ) Plaintiffs, ) ) v. ) Case No. 25-cv-03192 (APM) ) GOOGLE LLC, et al., ) ) Defendants. ) _________________________________________ ) MEMORANDUM OPINION I. INTRODUCTION Before the court are two further entries in a series of private antitrust lawsuits brought against Defendant Google LLC arising from its dominance in the market for general search services. Plaintiff Chegg, Inc. is an education technology company that provides students with online resources to improve their learning. Plaintiff Penske Media Corporation is a media, publishing, and information services company that owns multiple media properties and leading publications, several of which also join this suit (collectively “PMC”). 1 In their respective complaints, Plaintiffs assert that Google has leveraged monopoly power in the general search services market to coerce them and other online publishers to supply cost-free content to Google. Google then republishes that content to its users to compete with those same publishers, entrench its search monopoly, and expand that monopoly into different digital publishing markets. They bring near-identical actions against Google and its parent company Defendant Alphabet, Inc., alleging multiple violations of the Sherman Act and a common law claim of unjust enrichment. Defendants move to dismiss both actions on a variety of grounds. See Defs.’ Mot. to Dismiss Pl.’s Am. Compl., Chegg, Inc., v. Google, et al., No. 25-cv-543-APM (D.D.C.), ECF No. 19 [hereinafter Chegg Docket]; Defs.’ Mot. to Dismiss Pls.’ Am. Compl., Penske Media Corp., et al., v. Google, et al., No. 25-cv-3192-APM (D.D.C.), ECF No. 25 [hereinafter PMC Docket]. They argue, in sum, that Plaintiffs (1) fail to plead any coercive or anticompetitive arrangement that requires publishers to share their content with Google, (2) lack antitrust standing to assert claims related to the general search services market, (3) fail to plead relevant product markets or monopoly power in those markets, and (4) fail to support their unjust enrichment theory with the necessary factual allegations. For the reasons that follow, the court will grant the motions. II. BACKGROUND Because these matters come before the court on motions to dismiss, the court takes Plaintiffs’ well-pleaded factual allegations as true, Bell Atlantic Corporation v. Twombly, 550 U.S. 544, 555 (2007), and recites the factual background accordingly. 1 The PMC Plaintiffs include Penske Media Corporation and Plaintiffs Billboard Media, LLC, Deadline Hollywood, LLC, Fairchild Publishing, LLC, Gold Derby Media, LLC, Hollywood Reporter, LLC, Indiewire Media, LLC, Rolling Stone, LLC, SheMedia, LLC, and Variety Media, LLC. 2  Many of Plaintiffs’ allegations concerning Google’s general search engine, its dominance in the market for general search, and its generative artificial intelligence (“GenAI”) products are chronicled in United States v. Google LLC (Google Liability), 747 F. Supp. 3d 1 (D.D.C. 2024), and United States v. Google LLC (Google Remedies), 803 F. Supp. 3d 18 (D.D.C. 2025). 2 The court here will focus on the allegations unique to Plaintiffs’ claims. A. Digital Publishing The court begins with a discussion of the two main product markets 3 that are the subject of Plaintiffs’ complaints—Online Publishing and Online Educational Publishing—and then turns to a description of the alleged anticompetitive conduct. 1. Online Publishing According to the PMC Plaintiffs, “[t]he online publishing market consists of websites and apps on which publishers display textual content.” Am. Compl., PMC Docket, ECF No. 17 [hereinafter PMC Am. Compl.], ¶ 80. “Online publishing refers to the market for news articles, periodicals, reports, and other types of information that is made available online.” Id. Unlike content printed in physical form, “online publishing is consumed on digital devices connected to the Internet” and “can be accessed by anyone, anywhere, who has an Internet connection.” Id. ¶¶ 82–83. Digital publishers like Buzzfeed, The New York Times, and Condé Nast are participants in this market. Id. ¶ 85. So, too, are the PMC Plaintiffs. Founded over two decades ago, PMC is a “global media organization” that delivers content to more than 120 million monthly visitors in the United States 2 The court deems it appropriate to consider these decisions, as Plaintiffs’ Amended Complaints liberally reference the Google Liability proceedings, including the liability determination, trial exhibits and testimony, and the parties’ filings. See, e.g., Am. Compl., Chegg Docket, ECF No. 18 [hereinafter Chegg Am. Compl.], ¶¶ 47, 93, 94, 96, 125, 131, 138; Am. Compl., PMC Docket, ECF No. 17 [hereinafter PMC Am. Compl.]. ¶¶ 67, 73, 116, 118, 120, 121, 154, 171. 3 The geographic market for Plaintiffs’ claims is the United States. See Chegg Am. Compl. ¶¶ 54, 163; PMC Am. Compl. ¶¶ 80, 227. 3 through various digital properties, including several well-established entertainment and media outlets, such as Rolling Stone, Billboard, Variety, The Hollywood Reporter, and Deadline. Id. ¶¶ 32–40. Users generally are not required to pay to access PMC’s content. Id. ¶ 45. Instead, PMC generates revenue mostly through “digital advertising, commissions when a user purchases a product through an affiliate link on its sites, and subscription fees” for select products. Id. These revenue sources “depend on user traffic to PMC’s websites,” much of which comes through search engines like Google. Id. ¶¶ 45–46. 2. Online Educational Publishing Within digital publishing, Chegg alleges the existence of a distinct “educational publishing” market “for the production and dissemination of educational materials such as textbooks, workbooks, digital resources, and learning aids.” Am. Compl., Chegg Docket, ECF No. 18 [hereinafter Chegg Am. Compl.], ¶¶ 54, 56. Chegg refers to this as the “Online Educational Publishing” market. Id. ¶ 54. The content in this market is distinguished from other informational or non-fiction content by certain key attributes essential to student learning, including “curation, verification, authority, and pedagogical focus.” Id. ¶ 54. The “primary purpose” of this market is “to deliver information necessary for learning.” Id. ¶ 60. More specifically, the “content [is] designed primarily for structured learning, skill acquisition, or study,” which is “intended for repeated use over time and is periodically updated.” Id. The market is “targeted to learners seeking to supplement or obtain assistance with coursework, obtain academic support, or access learning products and services.” Id. ¶ 62. In the Online Educational Publishing market, Chegg is a “leading direct-to-student connected learning platform” that offers “personalized guidance” through subscription-based services to improve a student’s learning, using “prompts, learning tools, or assessment 4 opportunities based on the questions that student asks and the content they review on Chegg’s site.” Id. ¶¶ 23–24. For example, one of its subscription services, Chegg Study, provides “step- by-step learning support” and access to an “unparalleled [online] database of 135 million proprietary question-and-answer solutions.” Id. at 24. This digital content is Chegg’s “most valuable product,” id. ¶ 25, and Chegg has invested “hundreds of millions of dollars in the vast human capital and technological capabilities necessary” to create, maintain, and deliver its online content and services to subscribers, id. ¶ 26. 3. Digital Publishers and Search Engines Plaintiffs and other digital publishers “depend[] on search services” to distribute their content to online users. 4 Chegg Am. Compl. ¶ 29; PMC Am. Compl. ¶ 302. A user searching for an answer online enters their query into a search engine, which will then produce results on a search engine results page (“SERP”). Chegg Am. Compl. ¶ 32. From the SERP, a user can click on links that will take the user directly to publishers’ websites. See id. This is “the single-most important way” online publishers reach users and, by extension, generate revenue through user interactions with advertising, affiliate links, and subscriptions. Id. ¶¶ 32, 39; PMC Am. Compl. ¶ 66. To create search results, search engines like Google use algorithms that parse the content of their search indexes for what is relevant to a user’s query. Chegg Am. Compl. ¶ 34. A search index is a database containing copies of websites’ content and associated metadata, which Plaintiffs describe as “Search Index Data” for purposes of these lawsuits. Id. ¶ 32. “The search engine thus converts user attention to search referral traffic, which it ‘sells’ to the publisher 4 Plaintiffs’ allegations concerning how search engines operate and their interaction with online publishers are largely duplicative. Thus, while the court generally cites to Chegg’s Amended Complaint for conciseness, the PMC Plaintiffs make these same allegations. 5 (‘Search Referral Traffic’).” Id. Publishers “pay” for this traffic by contributing Search Index Data to search engines like Google, “so that the search engine can use that content to generate search results.” Id. In this way, Plaintiffs say, search engines are merely “intermediar[ies] between users seeking information and web publishers, who provide that information.” Id. ¶ 30 (emphasis omitted). Their purpose is “not to serve content” but instead to “connect users to where that content resides online.” Id. Publishers who “contribute high-quality content” to Google’s Search Index “are rewarded with search traffic.” Id. ¶ 38. Publishers can contribute Search Index Data in two ways: (1) allowing Google to use its “Googlebot” web crawler to visit their websites and collect their contents and related information; and/or (2) “pushing” their data directly to Google’s search index. See id. ¶¶ 35, 37. Both methods allow publishers to control what Google receives from them. Under the former, publishers can “opt out of Google’s search distribution and prevent their websites from appearing in Google’s search results” by editing a file on their websites that restricts what pages or sections of a website a web crawler can access. Id. ¶ 36. If a publisher does not do so, Google can crawl all or a portion of the publisher’s online content and include it in Google’s search index. Id. Under the latter method, publishers can choose what they transmit to Google, often sending their most current content to attract users. Id. ¶ 37. Plaintiffs share their content with Google with “the expectation that Google’s SERPs will direct users to” their websites. Chegg Am. Compl. ¶ 41; PMC Am. Compl. ¶ 68. And “when users click on a search result to visit [their] site[s],” publishers can “monetize that traffic.” Chegg Am. Compl. ¶ 41; PMC Am. Compl. ¶ 68. To Plaintiffs and other online publishers, this “quid pro quo”—their Search Index Data in exchange for Search Referral Traffic from Google—is the 6 “fundamental bargain” that undergirds the commercial internet. Chegg Am. Compl. ¶¶ 4, 33; PMC Am. Compl. ¶ 5. B. Google’s Appropriation of Publisher Content Plaintiffs claim that, in recent years, Google has unilaterally changed the terms of this “fundamental bargain.” See Chegg Am. Compl. ¶ 5; PMC Am. Compl. ¶ 6. Google previously did not compete with Plaintiffs in the relevant product markets. See generally PMC Am. Compl. ¶ 6; Chegg Am. Compl. ¶ 65. But starting in the early 2010s, Google “enter[ed] digital publishing by distributing content directly on its SERP.” Chegg Am. Compl. ¶ 63. It did so not by “hiring writers and editors,” but by “repurposing the content that digital publishers had created.” Id. This “appropriation” of content occurred in two phases involving different publishing elements on the SERP: (1) featured “snippets” of publishers’ content and (2) AI-generated narratives. Id. ¶ 64. 1. Featured Snippets In the first phase, Google began displaying detailed excerpts, or “snippets,” of publishers’ content on its SERP, beginning with news features. Id. ¶¶ 64, 66–69. Over time Google expanded its use of snippets to provide other types of content directly on the SERP. In May 2012, for example, Google introduced a “Knowledge Panel” that contains answers to different types of user queries. Id. ¶ 70. Knowledge Panels “obviate the need for users to leave the SERP page and click Google’s search result links to obtain answers to their questions.” Id. Google also began to include longer extracts of journalistic or informational articles on the Knowledge Panels, later termed “Featured Snippets.” Id. ¶ 71. Google continued to expand informational content on the SERP with other features. In 2015, it introduced a “People Also Ask” panel that “contains a list of questions about a user’s search topic, with drop-downs containing Featured Snippets chosen by Google to answer those 7 specific questions.” Id. ¶ 73. This feature “pull[s] out” specific parts of a publisher’s content to answer the searcher’s question. Id. ¶¶ 73–74. In addition, Chegg asserts that Google “targets” educational content through a “Questions and Answers” Featured Snippets format that excerpts questions and answers from Chegg and other websites. Id. ¶ 76. Google places these snippets ahead of organic search results on its SERP. Chegg Am. Compl. ¶ 124; PMC Am. Compl. ¶ 158. Plaintiffs say that these content features “are separate and distinct products from search results.” Chegg Am. Compl. ¶ 77. With Google’s introduction of these new SERP components, publishers faced a dilemma. Google “sources the content it uses to populate its [SERP] publishing elements from the data that it crawls for its search index.” Id. ¶ 80. A publisher could opt out of having its content republished as snippets on the SERP, but doing so would “prevent snippets from being shown as previews in search results.” Id. ¶ 82. The placement of Featured Snippets at the top of the SERP already reduced traffic to publisher websites. Opting out of snippets altogether would mean “an even greater reduction in search referrals,” as their content would not be as readily visible to users. Id. ¶¶ 81–82. “The decision to opt out of republishing by disallowing snippets or withholding Search Index Data is [thus] a Hobson’s choice.” Id. ¶ 83. “Virtually no digital publishers can afford to take such drastic action, because withholding data from Google’s search index means demotion on the SERP or disappearing from Google’s organic search results entirely.” Id. Because “appearing prominently in Google’s SERP is an essential means of generating traffic and revenue for digital publishers,” opting out is not a genuine choice at all. Id. 2. AI Overviews and Other GenAI Products The second phase of Google’s content appropriation began around 2023 with the launch of its GenAI products. Id. ¶ 93. The foundation of these products are large language models, or 8 LLMs. LLMs are “trained” on “vast datasets of written material, allowing them to encode patterns and relationships between words and sentences” that the LLM can then use to “predict[] the most likely next word based on the patterns it has learned.” Id. ¶¶ 99–100. Google trains its LLMs, in part, on publisher content scraped from websites by Google’s web crawlers and compiled into its search index. Id. ¶¶ 99–104, 133, 135. In addition, Google “grounds” its LLMs. Grounding, also known as retrieval-augmented generation (“RAG”), is a process that involves connecting the LLM to relevant, external content online and paraphrasing that content using generative AI. Id. ¶¶ 105– 106. Google launched two new GenAI products based on these LLM technologies: (1) “Search Generative Experience,” now known as “AI Overviews”; and (2) “Gemini,” a standalone chatbot. Id. ¶ 91. Placed at the top of the SERP, id. ¶ 124, AI Overviews are “AI- powered snapshot[s] in response to user queries,” sometimes consisting of a “machine-made essay consisting of multiple paragraphs purporting to provide the information that a user is searching for generated by an AI model from the very same publisher content that the user otherwise might have visited to learn the answer.” Id. ¶¶ 91, 122–124. An AI Overviews can paraphrase search results from publishers’ websites without linking those sources in the AI Overviews panel. Id. ¶ 127. But even when links are provided, a user satisfied by the AI Overviews response “will have little reason to click them.” Id. ¶ 128. Google’s other GenAI product, Gemini, “generates and publishes ‘original’ content in response to certain prompts,” sometimes excluding links to third party content. Id. ¶ 114. Gemini also generates educational content, such as study tips and learning tools. Id. ¶ 115. Google’s GenAI products now compete with digital publishers. By “generat[ing] answers to questions about information that is included in their training data,” Google creates 9 content that users otherwise would have to obtain directly from publishers’ websites. See id. ¶¶ 91, 100, 129. Publishers have limited ability to prevent Google from using their content to develop its GenAI products. Google offers a tool known as “Google Extended,” which allows publishers to block Google from using their content to improve Google’s LLMs. But switching on Google Extended does not prevent Google from using publishers’ content to train its base models or generate RAG responses to search queries. Id. ¶¶ 137–138, 166. The only effective way for a publisher to prevent Google from using its content to improve its GenAI products is to block Google from crawling its website. But blocking Google’s crawlers would mean omission from Google’s Search Index and, correspondingly, from the SERP in response to user queries. Id. ¶ 83; PMC Am. Compl. ¶ 106. And absence from the SERP would be devastating, as it would halt the large volume of referral traffic on which publishers rely to earn revenue. Id. Again, this is no true choice at all. Google thus deprives publishers “of choice and control over [their] content, by conditioning [their] appearance in search results on [their] permitting Google to use [their] content to feed Google Search’s artificial intelligence features.” Chegg Am. Compl. ¶ 93. C. Google’s Search Dominance and Harms to Digital Publishers Plaintiffs contend that “[b]ut for” Google’s exercise of its monopoly power in general search services Google would pay publishers like Plaintiffs for their content. PMC Am. Compl. ¶ 9; Chegg Am. Compl. ¶ 93. If Google refused to pay, publishers “would limit or block Google from crawling their websites for any purpose.” PMC Am. Compl. ¶ 9. Instead, Google’s dominance in search means that its “web crawlers remain largely free to index the web,” collecting information that it can repurpose at no cost. Chegg Am. Compl. ¶ 111. 10  Google’s “leverage[ing]” of its market dominance in search, Plaintiffs allege, gives it an unfair competitive advantage. Id. ¶¶ 95–96. Unlike other entrants on the “competitive fringe” of search that “seek to merge search results into AI-generated answers,” such as OpenAI and Perplexity, Google does not pay for publisher content. Id. ¶ 95. That content has substantial value, id. ¶¶ 97–98, as illustrated by several “real-world” agreements, PMC Am. Compl. ¶ 128. Plaintiffs point to The New York Times entering into a licensing agreement with Amazon that permits Amazon to use its content to train its LLMs. Id. ¶ 129. And Chegg has executed content licensing deals with two technology companies resulting in millions of dollars in revenue, and is in discussions with other companies to strike additional licensing agreements. Chegg Am. Compl. ¶ 97. Consequently, Plaintiffs claim that Google is a “monopsonist”—a buyer with dominant market power—in the “input market for publisher content used for search results.” Id. ¶ 52. Google “uses that buying power to force digital publishers to give up access to their content without monetary compensation.” Id. “Google then itself acts as a publisher, either by republishing portions of other digital publishers’ content or by using GAI to summarize the content.” Id. This all works to make Google an “answer engine,” rather than a search engine. Chegg Am. Compl. ¶ 77; PMC Am. Compl. ¶ 100. This conduct threatens to upend Plaintiffs’ business models. Rather than using Google to navigate to websites like Plaintiffs’ to answer their search queries, users now increasingly consume publishers’ “abridged or derivative” content on Google’s SERP, “starv[ing] those publishers of traffic and revenue.” Chegg Am. Compl. ¶¶ 28, 52; PMC Am. Compl. ¶ 53. These content features produce “lower click-through rates to the original sources from which Google generates the 11 answers.” PMC Am. Compl. ¶ 8. Indeed, recent studies show significant declines in the click- through rate to websites appearing in organic search results. Chegg Am. Compl. ¶¶ 142–145, 148. Google’s anticompetitive conduct, Plaintiffs assert, will have dire consequences for digital publishing, as well as for the “overall quality and quantity of the information accessible on the Internet.” Id. ¶ 149. If Google continues to train and ground its LLMs “by copying the original content of publishers without permission or payment” and then “us[ing] that very content to generate outputs that divert users from original sources,” again with no payment, “the economic incentives necessary for the creation and publication of high-quality original content will evaporate.” Id. “Less content of poorer quality will reduce website traffic, resulting in reduced revenue, and thus less spending on content creation, spawning even less content of even poorer quality and even less revenue, and so on in a downward spiral for content creators and publishers alike.” Id. D. Plaintiffs’ Claims Much like their complaints’ allegations, Plaintiffs’ claims substantially overlap. Each set of Plaintiffs advance a host of violations of the Sherman Act and one common law claim of unjust enrichment. Counts I of Plaintiffs’ Amended Complaints charge reciprocal dealing in violation of Section 1 of the Sherman Act, 15 U.S.C. § 1. Chegg Am. Compl. ¶¶ 184–196; PMC Am. Compl. ¶¶ 255–267. According to Plaintiffs, Google unlawfully conditions the “sale of Search Referral Traffic” to Plaintiffs on their supply of three types of content at no cost for purposes unrelated to providing search results. Chegg Am. Compl. ¶ 186; PMC Am. Compl. ¶¶ 257–259. First, Plaintiffs must make available content that Defendants can republish through snippets (“Republishing Content”). Chegg Am. Compl. ¶ 155. Second, Plaintiffs must provide content for 12 Defendants to train their LLMs (“GAI Training Content”). Id. Third, Plaintiffs must offer content that Defendants can use, repackage, and republish via RAG (“RAG Content”). Id. Plaintiffs allege that “[c]ontent supplied for each of these uses constitutes a separate product sold in a separate relevant product market: (1) the Republishing Content market; (2) the GAI Training Content market; and (3) the RAG Content market.” Id.; see also PMC Am. Compl. ¶ 258. Plaintiffs also allege a distinct market for Search Referral Traffic, which they define as a market for delivering users to websites from search results. Chegg Am. Compl. ¶¶ 152–154. As a result of Google’s conduct, Plaintiffs maintain they are “paid less for the sale of Republishing Content, GAI Training Content, and RAG Content than [they] would have but for [Defendant’] conduct” and have “lost revenues as a result of Google diverting traffic from Plaintiff[s’] website[s] in the form of lost subscription revenue from users’ visits to [their] site[s].” Id. ¶ 194; see also PMC Am. Compl. ¶ 265. Counts II mirror Counts I, except they allege reciprocal dealing in violation of Section 2 of the Sherman Act, 15 U.S.C. § 2 (“Section 2”). Chegg Am. Compl. ¶¶ 197–207; PMC Am. Compl. ¶¶ 268–278. Again, Plaintiffs claim that Defendants use their monopoly power in the general search services market to condition the sale of Search Referral Traffic on Plaintiffs supplying Republishing Content, GAI Training Content, and RAG Content for free. Chegg Am. Compl. ¶¶ 199–201; PMC Am. Compl. ¶¶ 268–278. Through this anticompetitive conduct, Plaintiffs claim, Defendants have “acquired and maintained” monopoly power in general search services because “[f]orcing digital publishers to provide Republishing Content, GAI Training Content, and RAG Content for free effectively lowers Google’s costs.” Chegg Am. Compl. ¶ 202; PMC Am. Compl. ¶ 273. 13  Counts III assert claims of “unlawful monopoly leveraging” in violation of Section 2. Chegg Am. Compl. ¶¶ 208–214; PMC Am. Compl. ¶¶ 279–285. Chegg alleges that Defendants have leveraged their power in general search services to create an unfair competitive advantage in the market for Online Educational Publishing. Chegg Am. Compl. ¶ 209. The PMC Plaintiffs aver the same as to the market for Online Publishing. PMC Am. Compl. ¶¶ 280–281. Plaintiffs maintain that Google’s conduct has caused digital publishers to go out of business or lay off staff, resulting in restricted output and reduced quality in these markets. Chegg Am. Compl. ¶ 210; PMC Am. Compl. ¶ 281. Counts IV advance claims of unlawful monopolization in violation of Section 2. Chegg Am. Compl. ¶¶ 215–221; PMC Am. Compl. ¶¶ 286–292. Specifically, Plaintiffs allege that by forcing them to provide content at no cost for training and grounding Google’s GenAI models and republishing, Defendants have willfully acquired and maintained monopoly power in the general search services market. Chegg Am. Compl. ¶ 216; PMC Am. Compl. ¶ 287. Count V of Chegg’s Amended Complaint and Count VI of the PMC Plaintiffs’ Amended Complaint assert claims of attempted monopolization in violation of Section 2. Chegg Am. Compl. ¶¶ 222–229; PMC Am. Compl. ¶¶ 307–314. Plaintiffs aver that Defendants engaged in the anticompetitive conduct described in the other counts with the “specific intent” of creating monopolies in digital publishing markets. Chegg Am. Compl. ¶ 224; PMC Am. Compl. ¶ 309. Chegg’s claim concerns the Online Educational Publishing market, Chegg Am. Compl. ¶ 224, while the PMC Plaintiffs’ concerns the online publishing market, PMC Am. Compl. ¶ 309. Count VI of Chegg’s Amended Complaint and Count VII of the PMC Plaintiffs’ Amended Complaint allege unjust enrichment in violation of California law. Chegg Am. Compl. ¶¶ 230– 231; PMC Am. Compl. ¶¶ 315–324. By using Plaintiffs’ content to train and ground Google’s 14 LLMs without compensation, Defendants have unlawfully enriched themselves at Plaintiffs’ expense. Id. Finally, only the PMC Plaintiffs assert an unlawful tying claim in Count V of their Amended Complaint. PMC Am. Compl. ¶¶ 293–306. They maintain that Google unlawfully ties AI Overviews to its general search product in violation of Section 2. Id. Plaintiffs allege myriad harms stemming from Defendants’ alleged conduct. Mainly, they claim lost revenues resulting from a decline in search traffic to their websites, see, e.g., Chegg Am. Compl. ¶ 149; PMC Am. Compl. ¶ 207, and from Google’s cost-free acquisition of their content, Chegg Am. Compl. ¶ 194; PMC Am. Compl. ¶ 265. They seek compensatory damages, restitution, and injunctive relief. Chegg Am. Compl. at 77; PMC Am. Compl. at 103. E. Procedural Background Chegg filed suit on February 24, 2025. Compl., Chegg Docket, ECF No. 1. Defendants then moved to dismiss, Defs.’ Mot. to Dismiss, Chegg Docket, ECF No. 16, after which Chegg filed an amended complaint on June 9, 2025, see Chegg Am. Compl. Defendants once again moved to dismiss on July 25, 2025. Defs.’ Mot. to Dismiss the Am. Compl., Chegg Docket, ECF No. 19 [hereinafter Defs.’ Chegg Mot.]. 5 The PMC litigation followed a similar path. Those plaintiffs filed suit on September 12, 2025. Compl., PMC Docket, ECF No. 1. Soon after, Defendants moved to dismiss the complaint, Defs.’ Mot. to Dismiss, PMC Docket, ECF No. 16, and the PMC Plaintiffs amended their pleading on December 4, 2025, see PMC Am. Compl. Defendants again moved to dismiss on January 12, 5 Defendants ask the court to take judicial notice of three documents while considering their motion to dismiss Chegg’s Amended Complaint. See Defs.’ Chegg Mot., Req. for Jud. Notice in Supp. of Defs.’ Mot. to Dismiss, Chegg Docket, ECF No. 19-2. Chegg opposes the request. See Pl.’s Mem. in Opp’n to Defs.’ Chegg Mot., Chegg Docket, ECF No. 20, Pl. Chegg’s Opp’n to Defs.’ Req. for Jud. Notice, ECF No. 20-1. As the court resolves the motion to dismiss without relying on the documents at issue, the court denies the motion. 15 2026. Defs.’ Mot. to Dismiss Pls.’ Amended Compl., PMC Docket, ECF No. 25 [hereinafter Defs.’ PMC Mot.]. The court agreed to a consolidated hearing on Defendants’ motions, Minute Order, PMC Docket, July 29, 2026, and held argument on August 25, 2026, see Hr’g Tr., PMC Docket, ECF No. 31 [hereinafter Consol. Hr’g Tr.]. III. LEGAL STANDARD To survive a motion to dismiss, a complaint must “state a claim to relief that is plausible on its face.” Ashcroft v. Iqbal, 556 U.S. 662, 678 (2009). The court must accept as true all well- pleaded factual allegations contained in the complaint, Twombly, 550 U.S. at 556, and construe the complaint in the plaintiff’s favor, Hettinga v. United States, 677 F.3d 471, 476 (D.C. Cir. 2012). But it must not do the same for “a legal conclusion couched as a factual allegation,” Papasan v. Allain, 478 U.S. 265, 286 (1986), nor should it “accept inferences drawn by [the] plaintiff if those inferences are not supported by the facts set out in the complaint,” Langeman v. Garland, 88 F.4th 289, 294 (D.C. Cir. 2023). In antitrust cases, the plaintiff must allege sufficient “fact[s] to raise a reasonable expectation that discovery will reveal evidence of illegal [conduct],” Twombly, 550 U.S. at 556, and the court must bear in mind that, while “it is one thing to be cautious before dismissing an antitrust complaint in advance of discovery,” it is “quite another to forget that proceeding to antitrust discovery can be expensive,” id. at 558 (internal citation omitted). Ultimately, the district court “must retain the power to insist upon some specificity in pleading before allowing a potentially massive factual controversy to proceed.” Id. 16 IV. DISCUSSION A. Reciprocal Dealing At the heart of Plaintiffs’ complaints are their reciprocal dealing claims (Counts I and II). So, the court starts there. Reciprocal dealing “exists where ‘two parties face each other as both buyer and seller.’” Brokerage Concepts, Inc. v. U.S. Healthcare, Inc., 140 F.3d 494, 511 (3d Cir. 1998) (quoting Spartan Grain & Mill Co. v. Ayers, 581 F. 2d 419, 424 (5th Cir. 1978)). In such an arrangement, the first party offers to buy the second party’s goods, but only if the second party buys other goods from the first party. Id. Put “more colloquially,” “reciprocal dealing exists when one party tells the other: ‘I’ll buy from you, if you buy from me.’” Id. Some courts have treated reciprocal dealing arrangements as a “variant” of tying arrangements—where a seller conditions the sale of one good on the buyer purchasing another, separate good—and have therefore analyzed these arrangements by similar standards. See Hicks v. PGA Tour, Inc., 897 F.3d 1109, 1115 (9th Cir. 2018); Spartan Grain, 581 F.2d at 425 (observing that tying arrangements and reciprocal dealing “refer to similar phenomena” in which “one side of a transaction has special power in the marketplace” and “uses this power to force those with whom it deals to make concessions in another market”); Brokerage Concepts, 140 F.3d at 512. That analogous treatment is not without its critics. Phillip E. Areeda & Herbert Hovenkamp, Antitrust Law ¶ 1776 (5th ed. & Suppl. 2026) [hereinafter Areeda & Hovenkamp]. Regardless, like tying, not all reciprocal dealing is anticompetitive, and the Sherman Act only proscribes reciprocal dealing that is “coercive,” such as “where a party uses its economic power as a purchaser in one market in order to restrict competition in another market where it is a seller.” Brokerage Concepts, 140 F.3d at 511 (citing Betaseed, Inc. v. U & I, Inc., 681 F.2d 1203, 1216 (9th Cir.1982)); see also FTC v. Consol. Foods Corp., 380 U.S. 592, 594–95 (1965). 17  Plaintiffs’ reciprocal dealing claims implicate four alleged product markets: (1) the Search Referral Traffic market; (2) the Republishing Content market; (3) the GAI Training market; and (4) the RAG Content market. Id. Because the particularities of the last three markets do not impact the court’s analysis, the court simply refers to them collectively as the “Content Inputs market.” Plaintiffs allege that Defendants condition “the sale of Search Referral Traffic” on publishers agreeing to surrender Content Inputs for free. Chegg Am. Compl. ¶¶ 186, 199; PMC Am. Compl. ¶¶ 257, 270. This, Plaintiffs contend, is a coercive reciprocal dealing arrangement that is illegal per se or, alternatively, under the rule of reason. Pl.’s Mem. in Opp’n to Defs.’ Chegg Mot., Chegg Docket, ECF No. 20 [hereinafter Pl.’s Chegg Opp’n], at 10–15; Pls.’ Mem. in Opp’n to Defs.’ PMC Mot., PMC Docket, ECF No. 26 [hereinafter Pls.’ PMC Opp’n], at 8–14. In response, Defendants argue there cannot be unlawful “reciprocal dealing” when, as here, there is no “deal.” Defs.’ Chegg Mot., Mem. of P. & A. in Supp. of Defs.’ Chegg Mot., ECF No. 19-1 [hereinafter Defs.’ Chegg Mem.], at 9; Defs.’ PMC Mot., Mem. of P. & A. in Supp. of Defs.’ PMC Mot., ECF No. 25-1 [hereinafter Defs.’ PMC Mot.], at 2. They maintain that Plaintiffs have failed to plead any actual agreement whereby Defendants promised to “sell” Plaintiffs any specific amount of traffic—or any traffic whatsoever—in exchange for “buying” their content. See Defs.’ Chegg Mem. at 9; see also Consol. Hr’g Tr. at 55:10-21. In Defendants’ telling, what Plaintiffs describe as coercion is a lawful refusal to deal on Plaintiffs’ preferred terms. Defs.’ Chegg Mem. at 10; Defs.’ PMC Mem. at 2. Defendants also argue that the alleged tied Content Input markets are themselves implausible and, even if they were to exist, Plaintiffs failed to allege any competitive harm to those markets. Defs.’ PMC Mem. at 15–20. The court agrees with Defendants that Plaintiffs have failed to allege plausible claims of reciprocal dealing. Regardless of whether those claims should be accorded per se treatment or 18 assessed under the rule of reason, or whether Plaintiffs have plead plausible markets, their reciprocal dealing claims fail to get out of the starting gate. 1. Section 1 Section 1 of the Sherman Act prohibits any “contract, combination . . . or conspiracy, in restraint of trade or commerce.” 15 U.S.C. § 1. To make out a claim under Section 1, plaintiffs must first allege that “the challenged anticompetitive conduct stems from . . . an agreement, tacit or express.” Twombly, 550 U.S. at 553 (cleaned up). “[S]tating such a claim requires a complaint with enough factual matter (taken as true) to suggest that an agreement was made.” Id. at 556; see also Donald F. Turner, The Definition of Agreement Under the Sherman Act: Conscious Parallelism and Refusals to Deal, 75 HARV. L. REV. 655, 655–56 (1962) (“For most kinds of anticompetitive business conduct, condemnation has depended and continues to depend on finding two or more parties who may be said to have ‘agreed’ to do what was done, since ‘agreement’ is an essential ingredient of ‘contract, combination, or conspiracy.’”). Plaintiffs thus must plead facts showing concerted action between independent actors, not unilateral conduct. See Copperweld Corp. v. Indep. Tube Corp., 467 U.S. 752, 767–69 (1984); Monsanto Co. v. Spray-Rite Serv. Corp., 465 U.S. 752, 764 (1984) (requiring evidence “that tends to exclude the possibility” of independent action). An agreement need not be in writing, as “[t]he antitrust laws look to substance, not to form.” Rothery Storage & Van Co. v. Atlas Van Lines, Inc., 597 F. Supp. 217, 228 (D.D.C. 1984), aff’d, 792 F.2d 210 (D.C. Cir. 1986); see also United States v. Parke, Davis & Co., 362 U.S. 29, 44 (1960) (“[J]udicial inquiry is not to stop with a search of the record for evidence of purely contractual arrangements” because the Sherman Act forbids agreements that “suppress competition,” “judged by what the parties actually did rather than by the words they used.”). But some meeting of the minds is required. See Contract, Black’s Law Dictionary (1ST ED. 1891) 19 (“A contract or agreement is either where a promise is made on one side and assented to on the other; or where two or more persons enter into [an] engagement with each other by a promise on either side.”); In re Baby Food Antitrust Litig., 166 F.3d 112, 117 (3d Cir. 1999) (“[U]nity of purpose or a common design and understanding or a meeting of the minds in an unlawful arrangement must exist to trigger Section 1 liability.”) (internal quotation marks omitted). Plaintiffs have not asserted a plausible “agreement” to maintain a Section 1 claim. They allege neither the essential elements of one nor the circumstances under which one arose. Plaintiffs do not assert that Defendants offered terms they accepted; the parties negotiated quantity, duration, or exclusivity; or either side made any commitment. They allege no communications between Plaintiffs and Google from which mutual assent to engage in a transaction could plausibly be inferred. Nor do they allege facts of a sudden change in pricing patterns or other conduct that could imply any agreement between these Plaintiffs and Google. See, e.g., Turner, The Definition of Agreement Under the Sherman Act, at 672–73. Not surprisingly, Plaintiffs have not cited any court that has recognized a reciprocal dealing claim when these basic facts are lacking. Plaintiffs have pleaded only that they have an “expectation” that Google will send them search traffic if they make their content available for free. Chegg Am. Compl. ¶ 41; PMC Am. Compl. ¶ 68. But an expectation is not an agreement. It is simply how a general search engine works. See Google Remedies, 803 F. Supp. 3d at 112–13. The court has previously rejected similar claims of an agreement between Google and publishers, and Plaintiffs offer no persuasive reason to change course. In Helena World Chron., LLC v. Google LLC, No. 23-cv-03677 (APM), 2026 WL 787882 (D.D.C. Mar. 20, 2026), plaintiffs who publish online news content indexed and distributed through Google Search and Google’s GenAI products claimed that Google engaged in a “monopoly broth” of anticompetitive acts, 20 including the same uncompensated scraping and republishing of publishers’ content that allegedly allowed Google to compete with other news publishers while starving them of search traffic and licensing revenue. Id. at *1. The court rejected the plaintiffs’ characterization of a “transactional arrangement with Google to supply news content in exchange for search traffic” because they did “not claim to have entered into any written or oral commercial agreement whereby they provide[d] money or news content to Google as consideration for search referrals; they allege[d] no actual agreement to purchase of any kind.” Id. at *8 (internal quotation marks omitted). 6 Like the Helena World Chronicle plaintiffs, Plaintiffs here offer no facts to suggest the existence of an agreement. And as in Helena World Chronicle, Plaintiffs’ mere assertion that an agreement exists amounts to a “legal conclusion couched as a factual allegation that the court is not bound to accept as true.” Helena World Chron., 2026 WL 787882, at *8 (quoting Papasan, 478 U.S. at 286). Plaintiffs argue that the court can find an “implied agreement” based on the “historical course of dealing between the parties.” Consol. Hr’g Tr. at 19:21–20:3; see also Pl.’s Chegg Opp’n at 1 (alleging Chegg has “a de facto agreement” with Google). The court acknowledges that the Supreme Court has recognized a “course of dealing” could evidence an implied agreement for purposes of the Sherman Act. See United States v. A. Schrader’s Son, 252 U.S. 85, 99 (1920) (describing “agreements—whether express or implied from a course of dealing or other circumstances—with all customers . . . to bind them to observe fixed resale prices”); see also Frey & Son v. Cudahy Packing Co., 256 U.S. 208, 210 (1921) (stating that Schrader’s Son “distinctly stated that the essential agreement, combination or conspiracy might be implied from a course of 6 Though the court then observed that the exchange of traffic for search index data was “characterized more accurately as a sale of one product conditioned on the reciprocal supply of another,” it did so only to mark a contrast with a classic tying arrangement that the plaintiffs had unsuccessfully alleged. Helena World Chron., 2026 WL 787882, at *13; Consol. Hr’g Tr. at 54:7-17. The court did not accept the notion that news publishers “sold” content to Google in exchange for search traffic. 21 dealing or other circumstance”). But such a course of dealing must at least plausibly supply the essential terms of the agreement, like the price or quantity of goods. See Frey & Son, 256 U.S. at 210–11. Plaintiffs do not plead such essential elements. The closest they come is to allege that they “permit[] Google to access their content for an access price of zero.” Chegg Am. Compl. ¶ 172. But Plaintiffs do not claim that this term arose from a meeting of the minds with Google, as opposed to their voluntarily acceding to cost-free crawling. The latter circumstance is the more plausible. See e.g., Llacua v. W. Range Ass’n, 930 F.3d 1161, 1181 (10th Cir. 2019) (“There is no allegation association members discussed or agreed among themselves how to pay foreign shepherds. Instead, the allegations in the [Complaint] simply indicate member ranches unilaterally decided to join the Association Defendants and utilize their services in filling out paperwork as they saw fit in their individual business judgment.”). Furthermore, the Supreme Court has already rejected the inference Plaintiffs would have the court draw from the mere duration and consistency of conduct: “an allegation of parallel conduct and a bare assertion of a[n] [agreement] will not suffice,” because parallel conduct is “just as much in line with a wide swath of rational and competitive business strategy” as any agreement. Twombly, 550 U.S. at 554, 556–57; see also Monsanto, 465 U.S. at 764. Although considered by the Court in the context of discerning a conspiracy, the principle applies with equal force here. Google’s crawlers index publicly available websites through an automated process. Publishers, for their part, keep their content open to crawling because doing so serves their own independent interest in search traffic, not because they struck any bargain with Google. Merely incanting the phrase “historical course of dealing” cannot plausibly convert two parties’ independently self- interested conduct into a contract. See Llacua, 930 F.3d at 1180–81. 22  The implausibility of Plaintiffs’ theory is underscored by its dramatic implications: if accepted, it would mean that Google contracts with every website that it indexes for Google Search. See Defs.’ Chegg Mem. at 9. Google therefore would have agreements with billions of web content publishers. Cf. Google Remedies, 803 F. Supp. 3d at 112–15 (explaining that Google crawls trillions of web pages, a fraction of which end up in the search index). That is entirely implausible, irrespective of whether the agreements are framed as “access” to traffic or a quantity of traffic. See Consol. Hr’g Tr. at 22:2-7. Perhaps recognizing this, Plaintiffs attempt to shrink this possible universe by pointing to a historical course of dealing with “publishers who create quality content,” since Google “drive[s] traffic toward those publishers” and “solicit[s]” those “publishers to create quality content and to optimize that content for Google’s Search index.” See id. at 20:19–21:8. But again, Plaintiffs allege no facts in support of an agreement, such as Google “charging” websites to index pages based on the “quality” of their content. And though it may be true that Google encourages publishers to create higher quality websites to drive traffic to their sites, Chegg Am. Compl. ¶ 38, PMC Am. Compl. ¶ 63, Google makes no promise that traffic will follow or in what quantity. That higher-quality websites generally attract more traffic is not a function of contract; once more, it is simply how general search engines work. See Google Remedies, 803 F. Supp. 3d at 112–13. In any event, distinguishing between digital publishers that produce “quality” content from those that do not is not an objective way to determine who has an “agreement” with Google for purposes of the Sherman Act. Judicial determinations of “quality” are ill-suited for this inquiry. 2. Section 2 Plaintiffs’ reciprocal dealing claims fare no better when brought under Section 2. While “Section 1 applies only to concerted action that restrains trade,” Section 2 “covers both concerted 23 and independent action, but only if that action monopolizes or threatens actual monopolization.” Am. Needle, Inc. v. Nat’l Football League, 560 U.S. 183, 190 (2010) (internal citations omitted and cleaned up). To plead a claim under Section 2, Plaintiffs must allege “the possession of monopoly power in the relevant market” and “the willful acquisition or maintenance of that power as distinguished from growth or development as a consequence of a superior product, business acumen, or historic accident.” Fed. Trade Comm’n v. Endo Pharms., 82 F.4th 1196, 1200 (D.C. Cir. 2023) (quoting United States v. Grinnell Corp., 384 U.S. 563, 570–71 (1966)). Having already set aside any plausible concerted action, independent action remains, and Plaintiffs suggest that the court can take a “broader view” of Google’s various anticompetitive acts under Section 2. See Consol. Hr’g Tr. at 26:10–27:5. But reciprocal dealing is, by definition, a two-sided practice: a defendant monopolist’s sale of a good must be linked to, or conditioned on, the counterparty’s reciprocal sale of another good. See Spartan Grain, 581 F.2d at 424. That is precisely how Plaintiffs frame their Section 2 claims. See Chegg Am. Compl. ¶¶ 199, 201; PMC Am. Compl. ¶¶ 270, 272 (alleging that Google “conditions the sale” of search referral traffic on Plaintiffs “giving Google” Content Inputs “for free”). While a monopolist’s coerced imposition of reciprocity on an otherwise unwilling counterparty can be exclusionary conduct, see Consol. Foods, 380 U.S. at 594–95, that still presupposes that a dealing relationship exists, see Spartan Grain, 581 F.2d at 424. And here, as explained above, Plaintiffs plead no circumstances that reveal “a unity of purpose or a common design and understanding” or a “meeting of minds.” See Am. Tobacco Co. v. United States, 328 U.S. 781, 810 (1946). Absent any actual reciprocal dealing, there can be no exclusionary conduct for purposes of Plaintiffs’ Section 2 claim. 24  * * * The court does not treat Plaintiffs’ alleged harms lightly. Nor is it unsympathetic to the situation publishers now find themselves in, and the knock-on consequences to journalists, educators, and other online creators whose content Google takes and repurposes without compensation. But the “antitrust statutes” are not “in any sense a substitute for [a] legislative body addressing questions of . . . economic dislocation caused by new innovation.” Areeda & Hovenkamp ¶ 100b. And “[t]o the extent that this case highlights any deficiencies” in the reach of the antitrust laws, “that is an issue for Congress [or regulators] to consider.” Citizens for Resp. & Ethics in Washington v. Cheney, 593 F. Supp. 2d 194, 198–99 (D.D.C. 2009). The court “is bound to apply the law only as it is written, not how the [c]ourt or any party believes it ought to be.” Id. at 199. “Because the [P]laintiffs here have not nudged their [reciprocal dealing] claims across the line from conceivable to plausible,” Twombly, 550 U.S. at 570, Counts I and II of their respective Amended Complaints must be dismissed. B. Tying The court turns next to the PMC Plaintiffs’ tying claim (Chegg makes no similar claim). PMC Am. Compl. ¶¶ 293–306. In their opposition to Defendants’ motion to dismiss, the PMC Plaintiffs chiefly argue that their tying claim should be subject to the per se test. Pls.’ PMC Opp’n at 35–40. The court assumes without deciding that the per se test applies and analyzes their claim accordingly. As previewed above, a tying arrangement is “an agreement by a party to sell one product but only on the condition that the buyer also purchases a different (or tied) product, or at least agrees that he will not purchase that product from any other supplier.” N. Pac. Ry. Co. v. United 25 States, 356 U.S. 1, 5–6 (1958). To establish the existence of an unlawful tie, a plaintiff must show that (1) an arrangement involves two (or more) separate products and (2) the seller “force[s] the buyer into the purchase of a tied product that the buyer either did not want at all, or might have preferred to purchase elsewhere on different terms.” Epic Games, Inc. v. Apple, Inc., 67 F.4th 946, 995 (9th Cir. 2023) (quoting Jefferson Par. Hosp. Dist. No. 2 v. Hyde, 466 U.S. 2, 12 (1984), abrogated on other grounds by Ill. Tool Works Inc. v. Indep. Ink, Inc., 547 U.S. 28 (2006)); see also United States v. Microsoft, 253 F.3d 34, 85 (D.C. Cir. 2001) (setting forth the elements of a per se tying claim, including “the tying and tied goods are two separate products” and “the defendant affords consumers no choice but to purchase the tied product from it”). Under the PMC Plaintiffs’ theory, general search is the tying product in which Google holds monopoly power, and AI Overviews is the separate tied product that Google forces internet users to “purchase” alongside it. PMC Am. Compl. ¶¶ 296–298. That coercion stems from the automatic appearance of AI Overviews at the top of the SERP, without any option for users to opt out. Id. ¶ 299. Moreover, Google’s Terms of Service bar users from modifying how its search services work. Id. Defendants counter that Google Search and AI Overviews are not separate products, as required to establish a tying arrangement. Defs.’ PMC Mem. at 34–36. The court again agrees with Defendants. The PMC Plaintiffs have not plausibly alleged that general search and AI Overviews are separate products, as opposed to an integrated search experience. “To constitute two separate products, ‘there must be sufficient consumer demand so that it is efficient for a firm to provide’ the products separately.” Epic Games, 67 F.4th at 995 (cleaned up) (quoting Eastman Kodak Co. v. Image Tech. Servs., 504 U.S. 451, 462 (1992); see also Microsoft, 253 F.3d at 86–88 (describing consumer demand test for separate products). 26 The inquiry “turns not on the functional relation between them, but rather on the character of the demand for the two items.” Jefferson Par., 466 U.S. at 19. As this court previously explained in Helena World Chronicle, “[i]n response to a given user query, Google’s SERPs may deliver many types of results, including organic links, ads, vertical offerings, and AI Overviews.” 2026 WL 787882, at *13 n.5. It is thus “hard to imagine consumer demand for one type of Google’s search results as separate from consumer demand for Google Search as a whole; demand for Google Search is inclusive of the demand for AI Overviews and any other item on Google’s SERPs.” Id. “If AI Overviews is a separate product from Google Search, then so is potentially every other type of search result Google delivers.” Id. The PMC Plaintiffs offer no persuasive reason to depart from this earlier conclusion. They support their argument for separate demand by characterizing Google Search and AI Overviews as having separate uses: “Search is meant to take a user from one place to another, the second place being the place that has the content that they want; and AI Overviews are designed to keep someone on the search engine results page.” Consol. Hr’g Tr. at 29:18-24; see also Pls.’ PMC Opp’n at 36. Their own allegations contradict that argument. If, as PMC Plaintiffs claim, there is separate consumer demand for Google Search to “travel to other webpages to explore information responsive to [their search] query,” then presumably Google’s introduction of AI Overviews at the top of the SERP would cause users to scroll past that feature to get to the organic links. See PMC Am. Compl. ¶ 56. But Plaintiffs allege just the opposite: that users are not scrolling past AI Overviews to get to the organic links. Instead, they contend, a user satisfied with the answer offered by AI Overviews would have “little reason” to navigate to a different website, id. ¶ 164, allowing Google to “cannibalize[]” the traffic intended for the PMC Plaintiffs, id. ¶ 187. These allegations suggest not separate consumer demands, but rather a general demand for information 27 and consumer indifference as to how they obtain it. Or put differently, the more plausible inference is that Google Search and AI Overviews have the same use case: providing answers to users’ search queries. In this way, the presence of AI Overviews on the SERP does not implicate the evil of tying arrangements, which is “prevent[ing] goods from competing directly for consumer choice on the merits . . . ” Microsoft, 253 F.3d at 87. For example, Plaintiffs make no factual allegation to support the notion that, if untied, users would elect Google Search to find and reach websites but opt for a different GenAI offering—such as from OpenAI or Anthropic—to supply a narrative summary at the top of a SERP. Further, Plaintiffs acknowledge that other general search engines, like DuckDuckGo, also offer a GenAI summary at the top of their SERPs. See Pls.’ PMC Opp’n at 37. “If a court finds either that there is no noticeable separate demand for the tied product or, there being no convincing direct evidence of separate demand, that the entire ‘competitive fringe’ engages in the same behavior as the defendant, then the tying and tied products should be declared one product and per se liability should be rejected.” Microsoft, 253 F.3d at 88 (citation omitted). Both circumstances apply here. By failing to plausibly allege that the “tying and tied goods are two separate products” and that Defendants “afford[] consumers no choice but to purchase the tied product” from them, no tying arrangement exists. The PMC Plaintiffs fail to make out the necessary elements of a per se tying claim. Microsoft, 253 F.3d at 85. Count V of their Amended Complaint is accordingly dismissed. C. Unlawful Monopoly Maintenance Plaintiffs’ unlawful monopoly maintenance claims are premised on Google’s alleged monopoly in the general search services market. Chegg Am. Compl. ¶ 216; PMC Am. 28 Compl. ¶ 287. To survive a motion to dismiss, Plaintiffs must plausibly establish that they have antitrust standing as to these claims. As with other plaintiffs appearing before this court, they have not. See Helena World Chron., 2026 WL 787882, at *6; see also Sensory, Inc. v. Google LLC, No. 24-cv-02788 (APM), 2026 WL 2017536, at *6 (D.D.C. July 13, 2026). A private antitrust plaintiff must establish antitrust standing. Andrx Pharms. v. Biovail Corp., 256 F.3d 799, 805–06 (D.C. Cir. 2001). Chief among the requirements of antitrust standing is antitrust injury. Id. at 812 (D.C. Cir. 2001) (citing 2 Phillip E. Areeda et al., Antitrust Law ¶ 337a (2d ed. 2000)). Antitrust injury is “injury of the type the antitrust laws were intended to prevent and that flows from that which makes defendants’ acts unlawful.” Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 489 (1977). The relevant “inquiry turns on whether the plaintiff is a participant in the relevant market and ‘suffered its injury in the market where competition is being restrained.’” Fotobom Media, Inc. v. Google LLC, 719 F. Supp. 3d 33, 44 (D.D.C. 2024) (quoting Am. Ad Mgmt., Inc. v. Gen. Tel. Co. of Cal., 190 F.3d 1051, 1057 (9th Cir. 1999)). Plaintiffs claim to be participants in the general search services market with antitrust standing under two theories. First, Plaintiffs fashion themselves as “suppliers” of Search Index Data in the market for general search services and “purchasers” of Search Referral Traffic. Pl.’s Chegg Opp’n at 33–34; Pls.’ PMC Opp’n at 31. Alternatively, and simultaneously, Plaintiffs argue they have standing because their injuries are “inextricably intertwined” with the injuries suffered by Google’s new AI competitors in the general search services market under Blue Shield v. McCready, 457 U.S. 465, 484 (1982). See Pls.’ Chegg Opp’n at 35; Pls.’ PMC Opp’n at 32–33. Neither theory persuades. 29  1. Market Participant Theory In rare cases, market participants other than consumers and competitors may suffer a cognizable antitrust injury. See Am. Ad Mgmt., 190 F.3d at 1057. But such participants still must be comparable to “a customer who obtains services in the threatened market or a competitor who seeks to serve that market.” SAS of P.R., Inc. v. P.R. Tel. Co., 48 F.3d 39, 44 (1st Cir. 1995). An injury that is “too secondary and indirect” from the anticompetitive conduct cannot establish antitrust standing. Serfecz v. Jewel Food Stores, 67 F.3d 591, 597 (7th Cir. 1995). Again, this court examined and dismissed this same theory of standing in Helena World Chronicle. There, the court dismissed the publisher-plaintiffs’ claims tied to the general search services market because their alleged injuries as “suppliers” of content or “purchasers” of traffic were too “secondary and indirect to be considered antitrust injuries.” Helena World Chron., 2026 WL 787882, at *6–8 (internal quotation marks omitted). That was in part because the publisher-plaintiffs’ claimed harms as suppliers were not harms suffered in the market for general search services. Id. at *7. Rather, they were revenue losses incurred in the alleged market for digital content. Id. As to their purchaser theory, it too was a non-starter because the publisher- plaintiffs failed to allege the existence of a purchasing agreement between them and Google. Id. at *8. And, at an even more basic level, it was unclear how the publisher-plaintiffs could allege to be both supplier and purchaser simultaneously: “If Plaintiffs ‘sell’ their content to Google in exchange for search traffic but also purport to buy this search traffic, then search traffic would be both the good purchased from Google and the consideration furnished to Google for that purchase. It cannot be both.” Id. at *7. The court’s skepticism is unchanged. Whether styled as suppliers or purchasers, Plaintiffs’ market participant theory fails for the same reasons, and the court need not spill much more ink 30 explaining why. As to their supplier theory, the alleged harms to Plaintiffs here do not arise in the market for general search services. They describe lost revenue from lost subscriptions and inadequate compensation for use of their content for training and grounding LLMs. These injuries, “though flowing from that which [allegedly] makes the defendant’s conduct unlawful, are experienced in another market.” FTC v. Qualcomm Inc., 969 F.3d 974, 992 (9th Cir. 2020). Parties suffering such injuries “do not suffer antitrust injury.” Id. As to their purchaser theory, as previously explained, Plaintiffs have failed to offer any well-pleaded facts to support the existence of an agreement to purchase traffic from Google. Consequently, Plaintiffs’ market participant theory fails to confer antitrust standing. 2. Standing Under Blue Shield v. McCready Plaintiffs’ arguments under McCready fare no better. In Blue Shield v. McCready, a health- plan subscriber was denied reimbursement for treatment by a psychologist under a plan that paid for the same services only when billed by a psychiatrist or through a physician—a policy Blue Shield adopted together with a psychiatric trade group to drive psychologists out of the psychotherapy business. 457 U.S. at 468–70. The Supreme Court reasoned that the plaintiff’s injury was “the very means by which it [was] alleged that Blue Shield sought to achieve its illegal ends,” such that her harm was not a byproduct of the conspiracy but an “integral” instrument of it. Id. at 479. McCready thus only reaches cases “where the plaintiff was ‘used as a conduit to harm the defendants’ actual competitors,’ such that the plaintiff’s harm is ‘an indispensable aspect of the scheme.’” Fotobom, 719 F. Supp. 3d at 47 (cleaned up) (quoting Hanover 3201 Realty, LLC v. Vill. Supermarkets, Inc., 806 F.3d 162, 173 (3d Cir. 2015)). Even then, “harm that is secondary to the anticompetitive conduct cannot support antitrust injury.” Hanover 3201 Realty, 806 F.3d at 173 (discussing McCready). 31  Plaintiffs argue that their injury is “inextricably intertwined under McCready because . . . obtaining content at zero cost is precisely the mechanism by which Google obtains the advantage in the new dynamic competition for search demand with generative AI.” Consol. Hr’g Tr. at 7:19– 8:2. More specifically, “[b]y obtaining [Plaintiffs’] content for republishing, training, and grounding at zero cost, Google effectively increases the relative costs to its competitors of obtaining content for these purposes.” Pls.’ PMC Opp’n at 33. Plaintiffs, however, have not plausibly alleged that their claimed harm is an “indispensable aspect” of Google’s alleged scheme. Their claim is that “[t]hrough” securing publisher content at no cost “Google has willfully acquired and maintained its monopoly power in [g]eneral [s]earch [s]ervices.” PMC Am. Compl. ¶ 287. But nowhere do Plaintiffs connect Google’s obtaining free content with its acquisition of dominant market share in general search. In fact, they make no allegations about market share at all. Probably for good reason. This court found in Google Remedies that, by 2009, 80% of all general search queries flowed through Google; by 2020, that percentage had increased to 89.2%. 747 F. Supp. 3d at 95. Plaintiffs do not allege that, before or during this period, among general search firms only Google acquired content free of cost. If the competition likewise was crawling and indexing websites at no cost to generate search results— and Plaintiffs do not contend otherwise—then Plaintiffs would have suffered no injury at all in connection with Google’s acquisition of monopoly power. The playing field would be level in that respect. Nor have Plaintiffs alleged harm that is indispensable to Google maintaining its monopoly in search. Google’s use of publisher content to train LLMs and for grounding to create GenAI responses to search queries is a recent phenomenon. PMC Am. Compl. ¶¶ 149–156 (alleging that Google rolled out Search Generative Experience, a precursor to AI Overviews, for the first time in 32 May 2023 “in an experimental phase”; AI Overviews was made available to all U.S. users in August 2024). Even if some general search competitors or nascent competitors from outside the search market (e.g., OpenAI) have had to pay for content during this more recent period, and Google has not, Plaintiffs offer no plausible theory for how Google’s alleged exclusionary conduct has allowed Google to maintain its monopoly in general search, beyond making it more costly for these firms to license AI training data. See Microsoft, 253 F.3d at 79. Plaintiffs do not assert that all or most of Google’s search competitors pay to license training data and that Google is the sole outlier, or that these costs, even if incurred, are prohibitive and cause anticompetitive effects. See PMC Am. Compl. ¶¶ 223–224. Plaintiffs’ loss of revenue, see id. ¶ 290, cannot be an “indispensable part of [Google’s] scheme” based on their mere say so, see Fotobom, 719 F. Supp. 3d at 47. Finally, Plaintiffs seek refuge in Novell, Inc. v. Microsoft Corp., 505 F.3d 302 (4th Cir. 2007). There, the plaintiff, Novell, alleged that Microsoft engaged in anticompetitive conduct to protect its monopoly in PC operating systems, even though Novell itself competed in the downstream applications market. Id. at 308. The Fourth Circuit determined that Novell was “a member of a limited class of plaintiffs” who possessed standing despite being outside the restrained market. Id. at 320. Applying the five factors from Associated General Contractors of California, Inc. (AGC) v. California State Council of Carpenters, 459 U.S. 519 (1983), the Fourth Circuit found that Microsoft’s conduct was aimed at Novell in particular: internal Microsoft communications showed the company deliberately timed and calibrated its conduct toward Novell’s products by name, and the challenged practice—withholding information Novell specifically needed, and contracting with distributors in ways that disadvantaged Novell’s offering relative to Microsoft’s own—were, on the pleaded facts, designed with Novell as their intended 33 target, rather than as an incidental effect of conduct aimed at the market at large. Id. at 315–20. These factors, in part, “weigh[ed] in favor of finding Novell to be the most direct victim with incentive to serve as a private attorney general.” Id. at 319. In Plaintiffs’ telling, Microsoft’s specific targeting of Novell is “very, very similar” to their claim that, “by acquiring content that its nascent AI competitors would have to pay for free, Google is obtaining a structural advantage in competition in the market for general search services.” Consol. Hr’g Tr. at 8:3-25. Novell bears no resemblance to this case. Plaintiffs offer no allegations that Google’s conduct was directed at them specifically or that Google even viewed itself as competing with them. Cf. Novell, 505 F.3d at 317 (citing evidence of “Microsoft Chairman Bill Gates specifically suggest[ing] waiting to publish critical technical specifications of Windows 95 until ‘[they had] a way to do a high level of integration [between Microsoft Office and Windows 95] that [would] be hard for [the] likes of . . . [Novell’s product] to achieve’” because, otherwise, Microsoft could not “‘compete with . . . [Novell]’”). Indeed, Plaintiffs come nowhere close to identifying anything comparable to the facts of individualized targeting and harm that persuaded the Fourth Circuit to treat Novell as Microsoft’s direct and intended victim. “At most, [Plaintiffs’] alleged injur[ies] [are] no more than ‘incidental’ or a ‘byproduct’ of Google’s anticompetitive conduct” in the general search services market. See Sensory, 2026 WL 2017536, at *7 (quoting In re Aluminum, 833 F.3d 151, 161 (2d Cir. 2016)). * * * Antitrust standing must be established for an antitrust claim to survive. See Johnson v. Comm’n on Presidential Debates, 869 F.3d 976, 981 (D.C. Cir. 2017). Because Plaintiffs have not pleaded sufficient facts to establish antitrust standing in the general search services market, the court dismisses Counts IV of Plaintiffs’ Amended Complaints. 34  D. Attempted Monopolization and Monopoly Leveraging Though Plaintiffs allege separate monopoly leveraging and attempted monopolization claims, the court addresses them together for reasons discussed below. For their leveraging claim, Plaintiffs allege that Defendants have used their monopoly power in general search services to gain an unfair competitive advantage in the Online Publishing and Online Educational Publishing markets. PMC Am. Compl. ¶¶ 280–281; Chegg Am. Compl. ¶¶ 60, 209. Simply put, each set of Plaintiffs allege that Defendants are using their dominance in one market to gain an unfair advantage in a second, unrelated market. Plaintiffs’ attempted monopolization claims differ from their leveraging claims in one key respect. The latter requires a showing of mere unfair advantage in the second market whereas the former requires proof that the anticompetitive conduct creates a “dangerous probability” of Defendants’ acquiring monopoly power in the second market. PMC Am. Compl. ¶¶ 307–314; Chegg Am. Compl. ¶¶ 222–229; see Microsoft, 253 F.3d at 80 (setting forth the elements of attempted monopolization). The vitality of a standalone monopoly leveraging claim under the Sherman Act is doubtful. See Areeda & Hovenkamp ¶ 652b. In Spectrum Sports v. McQuillan, 506 U.S. 447 (1993), the Supreme Court seemingly cast aside “nonmonopolistic leveraging” claims, like those asserted by Plaintiffs, under Section 2. It observed that “Congress authorized Sherman Act scrutiny of single firms only when they pose a danger of monopolization. Judging unilateral conduct in this manner reduces the risk that the antitrust laws will dampen the competitive zeal of a single aggressive entrepreneur. Thus, the conduct of a single firm, governed by [Section] 2, is unlawful only when it threatens actual monopolization.” Spectrum Sports, 506 U.S. at 456 (internal quotation marks and citation omitted). The Court later commented that “the notion that proof of unfair or predatory conduct alone is sufficient to make out the offense of attempted monopolization is contrary to the 35 purpose and policy of the Sherman Act.” Id. at 457. Then, in Verizon Communications Inc. v. Law Offices of Curtis V. Trinko, LLP, 540 U.S. 398 (2004), the court criticized the lower court for embracing a less stringent “monopoly leveraging” theory. Citing Spectrum Sports, the Court stated that, “[t]o the extent the Court of Appeals dispensed with the requirement that there be a ‘dangerous probability of success’ in monopolizing a second market, it erred.” Id. at 415 n.4 (citing Spectrum Sports, 506 U.S. at 459). Still, as Plaintiffs point out, one court in this District has recognized a monopoly leveraging claim post-Trinko. See 2301 M Cinema LLC v. Silver Cinemas Acquisition Co., 342 F. Supp. 3d 126, 133 (D.D.C. 2018) (recognizing a monopoly leveraging claim where plaintiffs alleged that the defendant “leverage[d] its monopoly power” by coercing market plaintiffs to accept agreements that favor the defendant). Whatever embers may remain of a nonmonopolistic leveraging claim, this court need not try to discern them. Both monopoly leveraging and attempted monopolization claims require the plaintiff to plausibly plead anticompetitive effects in a second distinct market. See Virgin Atl. Airways Ltd. v. Brit. Airways PLC, 257 F.3d 256, 272–73 (2d Cir. 2001); Microsoft, 253 F.3d at 81. Plaintiffs have failed to carry their burden. Defining a relevant market “establishes a context for evaluating a defendant’s actions as well as for measuring whether the challenged conduct presented a dangerous probability of monopolization.” Microsoft, 253 F.3d at 81. A “relevant market consists of all products reasonably interchangeable by consumers for the same purposes.” PhantomALERT Inc. v. Apple Inc., 183 F.4th 790, 800 (D.C. Cir. 2026) (internal quotation marks omitted). It contains both a “geographic” component and a “product” component.” Id. To define the latter, “antitrust plaintiffs may resort to tools aimed at measuring the ‘cross- elasticity of demand,’ designed as the ‘degree to which a similar product will be substituted for 36 the product in question.’” Id. (quoting Rothery Storage & Van Co., 792 F.2d at 218). Plaintiffs may also draw on “various ‘practical indicia’ of a relevant product market,” id. at 801, as outlined in Brown Shoe Co. v. United States, 370 U.S. 294 (1962). These factors include “industry or public recognition of the submarket as a separate economic entity, the product’s peculiar characteristics and uses, unique production facilities, distinct customers, distinct prices, sensitivity to price changes, and specialized vendors.” Brown Shoe Co., 370 U.S. at 325. The D.C. Circuit has “described the Brown Shoe factors as ‘evidentiary proxies for direct proof of substitutability.’” PhantomALERT Inc., 183 F.4th at 801. Though this is ordinarily a fact-intensive inquiry unsuited for resolution prior to discovery, an antitrust plaintiff must still offer “‘more than labels and conclusions’ or a ‘formulaic recitation of the elements of a cause of action.’” Id. (quoting Twombly, 550 U.S. at 555). 1. Online Publishing Starting with “Online Publishing,” the PMC Plaintiffs define that product market as encompassing essentially all digital, text-based content published online. See PMC Am. Compl. ¶ 80. It excludes only print media, while treating other categories of online content as reasonably interchangeable substitutes for another. Id. ¶¶ 82–85. The only “peculiar characteristic” of the market that they offer is that it can be consumed on any digital device, as opposed to in hard-copy form. Pls.’ PMC Opp’n at 23; Consol. Hr’g Tr. at 33:8-12. Such market definition is implausibly overbroad. The PMC Plaintiffs fail to explain why a blog post, a legal brief, a fictional story, or a news article—or, even, Chegg’s vast repository of questions and answers—can reasonably be considered substitutes in a single market. Nor do they attempt to define the market through the Brown Shoe factors or by any other means. 7 See Pls.’ 7 The PMC Plaintiffs apply the Brown Shoe factors to support defining the Search Referral Traffic and Content Inputs markets, but not the Online Publishing Market. See Pls.’ PMC Opp’n at 18–23. 37 PMC Opp’n at 23. A market that consists of “news articles, periodicals, reports and any other types of information that is made available online,” PMC Am. Compl. ¶ 80 (emphasis added), is simply not plausible. What’s more, this market definition, even if the court were to accept it, renders implausible their attempted monopolization theory. See id. ¶ 310. Indeed, the PMC Plaintiffs do not even attempt to quantify Google’s market share in online publishing or offer a single fact to support their allegation that Google has a dangerous probability of monopolizing the sweeping market they describe. Cf. Helena World Chron., 2026 WL 787882, at *9–10. Nor is it clear how the PMC Plaintiffs could allege that Google could restrict output or erect market barriers if any individual can create text-based digital content at any time and enter the online publishing market. See United States v. Syufy Enters., 903 F.2d 659, 664 (9th Cir. 1990) (“Time after time, we have recognized this basic fact of economic life: A high market share, though it may ordinarily raise an inference of monopoly power, will not do so in a market with low entry barriers or other evidence of a defendant’s inability to control prices or exclude competitors.”). 2. Online Educational Publishing Market Chegg defines the Online Educational Publishing Market as one “for the production and dissemination of educational materials such as textbooks, workbooks, digital resources, and learning aids” that can be accessed online. Chegg Am. Compl. ¶¶ 54, 56, 60. Such content is distinguished from other informational or non-fiction content by certain key attributes required for student learning, such as “curation, verification, authority, and pedagogical focus.” Id. ¶ 54. And it is “intended for repeated use over time and is periodically updated,” and is “targeted to learners seeking to supplement or obtain assistance with coursework, obtain academic support, or access learning products and services.” Id. ¶¶ 60, 62. 38  Unlike the PMC Plaintiffs, Chegg has at least attempted to offer some bounded market, and fleetingly addresses the Brown Shoe factors when defining the market for Online Educational Publishing. Pl.’s Chegg Opp’n at 38. Still, the product market Chegg puts forward remains hopelessly vague. To describe the market’s peculiar uses and distinct customers, for example, Chegg points to allegations that distinguish online educational publishing content from other online content by explaining that “[t]he primary purpose of Online Educational Publishing is pedagogical: to deliver information necessary for learning.” See, e.g., Chegg Am. Compl. ¶ 60; see also id. ¶ 54 (identifying “pedagogical focus” as a “key attribute[]” of the Online Educational Publishing market). It is unclear, however, how the court can discern whether a product’s “primary purpose” is pedagogical, as opposed to merely informational. Chegg suggests that other “key attributes,” such as “curation, verification, [and] authority,” can distinguish educational publishing content from other nonfiction content, but it is not clear how that is so. Id. ¶ 54; Pl.’s Chegg Opp’n at 39. Indeed, Chegg fails to explain how content produced by news and other nonfiction publishers is not also curated and verified. See Chegg Am. Compl. ¶ 54. Without more factual specificity, the court cannot reasonably assess substitutability or the practical indicia outlined in Brown Shoe. This claim fails for another reason: it is not entirely clear that Google is even a participant in the market for online educational publishing that Chegg describes. If the “primary purpose” of Google providing answers to users’ search queries is pedagogical, then any informational or factual content online could conceivably be swept into the market for online educational publishing, collapsing the market’s boundaries altogether. Moreover, Chegg offers no facts to support finding that the content that Google publishes in AI Overviews is “curat[ed]” or “verif[ied].” Chegg Am. Compl. ¶ 54. Indeed, its allegations are to the contrary. Chegg alleges 39 that Google LLMs operate by scraping online education publishers’ content, training on such content, and then using it to deliver answers to users’ search queries. Id. ¶¶ 101, 124. There is no curation or verification described in that process at all. Therefore, even if the court accepts that Chegg has defined a market for online educational publishing, Chegg’s claim would still fail because it has not plausibly pleaded that Google has a dangerous probability of monopolizing a market in which it participates. * * * By failing to define plausible markets, Plaintiffs do not make out a necessary element of their attempted monopolization claims. See Microsoft, 253 F.3d at 81. Their monopoly leveraging claims fall for the same reason. Both claims also suffer from other basic pleading deficiencies. Count V of Chegg’s Amended Complaint and Count VI of the PMC Plaintiffs’ Amended Complaint must be dismissed. E. Unjust Enrichment Finally, Plaintiffs bring state law claims for unjust enrichment. Chegg Am. Compl. ¶¶ 230–239; PMC Am. Compl. ¶¶ 315–324. All parties agree that California state law governs but differ as to its application. Defs.’ Chegg Mem. at 41–43; Defs.’ PMC Mem. at 42–44; Pl.’s Chegg Opp’n at 43; Pls.’ PMC Opp’n at 44. Having dismissed all federal claims in these actions, the court has discretion to either exercise supplemental jurisdiction over the remaining state law claims or decline to do so. See 28 U.S.C. § 1367(c); Matthews v. Dist. of Columbia, 507 F. Supp. 3d 203, 212 (D.D.C. 2020) (dismissing all federal law claims and declining to exercise supplemental jurisdiction over remaining state law claims, including unjust enrichment). In exercising this discretion, the court considers “judicial economy, convenience, and fairness to litigants.” United Mine Workers of Am. 40 v. Gibbs, 383 U.S. 715, 726 (1966). In the typical case in which all federal-law claims are dismissed, the “‘balance of factors to be considered . . . will point toward declining to exercise jurisdiction over the remaining state-law claims.’” Anderson v. Holder, 647 F.3d 1165, 1174 (D.C. Cir. 2011) (quoting Carnegie-Mellon Univ. v. Cohill, 484 U.S. 343, 350 n.7 (1988)). After careful consideration of the relevant factors, see 28 U.S.C. § 1367(c), the court declines to exercise supplemental jurisdiction over Plaintiffs’ unjust enrichment claims. Accordingly, the court will dismiss Chegg’s Count VI and the PMC Plaintiffs’ Count VII. V. CONCLUSION For the foregoing reasons, Defendants’ Motion to Dismiss Plaintiff Chegg’s Amended Complaint is granted. Chegg Docket, ECF No. 19. Defendants’ Motion to Dismiss the PMC Plaintiffs’ Amended Complaint is also granted. PMC Docket, ECF No. 25. A final, appealable order accompanies this Memorandum Opinion. Dated: September 30, 2026 Amit P. Mehta United States District Judge 41