NetChoice, LLC v. David Yost

• Court: Court of Appeals for the Sixth Circuit
• Decided: June 18, 2026
• Docket: 25-3371
• Status: Published

Opinion

RECOMMENDED FOR PUBLICATION Pursuant to Sixth Circuit I.O.P. 32.1(b) File Name: 26a0177p.06

UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT

┐ NETCHOICE, LLC, │ Plaintiff-Appellee, │ > No. 25-3371 │ v. │ │ DAVE YOST, in his official capacity as Ohio Attorney │ General, │ Defendant-Appellant. │ ┘

Appeal from the United States District Court for the Southern District of Ohio at Columbus. No. 2:24-cv-00047—Algenon L. Marbley, District Judge.

Argued: February 4, 2026

Decided and Filed: June 18, 2026

Before: BATCHELDER, CLAY, and RITZ, Circuit Judges.

_________________

COUNSEL

ARGUED: Mathura J. Sridharan, OFFICE OF THE OHIO ATTORNEY GENERAL, Columbus, Ohio, for Appellant. Erin E. Murphy, CLEMENT & MURPHY, PLLC, Alexandria, Virginia, for Appellee. ON BRIEF: Mathura J. Sridharan, Zachery P. Keller, OFFICE OF THE OHIO ATTORNEY GENERAL, Columbus, Ohio, for Appellant. Scott A. Keller, LEHOTSKY KELLER COHN LLP, Washington, D.C., for Appellee. Kevin A. Golembiewski, OFFICE OF THE FLORIDA ATTORNEY GENERAL, Tallahassee, Florida, James Emory Smith, OFFICE OF THE ATTORNEY GENERAL OF SOUTH CAROLINA, Columbia, South Carolina, Robert Corn-Revere, D Gill Sperlein, FOUNDATION FOR INDIVIDUAL RIGHTS AND EXPRESSION, Washington, D.C., Aaron Mackey, ELECTRIC FRONTIER FOUNDATION, San Francisco, California, for Amici Curiae.

CLAY, J., announced the judgment of the court and delivered the lead opinion. BATCHELDER, J. (pp. 34–60), concurred in the judgment and delivered a separate concurring opinion. RITZ, J. (pp. 61–71), delivered a separate dissenting opinion. No. 25-3371 NetChoice, LLC v. Yost Page 2

OPINION _________________

CLAY, Circuit Judge. Defendant David Anthony Yost, the Attorney General of Ohio, appeals from the district court’s final order and judgment in Plaintiff NetChoice, LLC’s civil suit challenging Ohio’s Parental Notification by Social Media Operators Act, H.B. 33, 135th Gen. Assemb., Reg. Sess. (2023) (codified at Ohio Rev. Code § 1349.09), on First Amendment and vagueness grounds, under 42 U.S.C. § 1983 and 28 U.S.C. § 2201. Because a majority of the panel agrees that NetChoice has failed to establish that the Act is facially unconstitutional, we REVERSE the district court’s judgment and REMAND with instructions to enter judgment in favor of Yost.

I. BACKGROUND

The state of Ohio is concerned that social media is harming Ohio’s youth and has sought to target that harm through legislative policy. Specifically, Ohio has taken notice of a growing body of evidence linking social media to poor mental health, eating disorders, and academic decline in youth and adolescents. The state further worries about the prevalent use of social media among child sexual predators to target minors, deficient data privacy for minor social media users, and exploitative contract terms that social media operators impose on them. Meanwhile, the members of Plaintiff trade association NetChoice, LLC (NetChoice), that own and operate social media platforms, extoll the virtues of social media and the benefits that it offers young people, including community-building, artistic expression, education, civic engagement, awareness of news and current events, and career development. Those entities tend to resist the legislative efforts of Ohio and other states to curb social media uptake among minors, asserting that such restrictions violate social media users’ and providers’ First Amendment rights. See NetChoice v. Carr, 789 F. Supp. 3d 1200, 1212 (N.D. Ga. 2025) (“A handful of states have enacted [] laws aimed at protecting minors online . . . . NetChoice has challenged at least eight other state laws[, and n]early all of those state laws [have been] enjoined . . . .”). Defendant David Yost (Yost), the Attorney General of Ohio, defends Ohio’s Parental Notification by Social Media Operators Act, H.B. 33, 135th Gen. Assemb., Reg. Sess. (2023) No. 25-3371 NetChoice, LLC v. Yost Page 3

(codified at Ohio Rev. Code § 1349.09) (the Act), as a legitimate exercise of the state’s prerogative to regulate contracting with minors. To the extent that the Act does burden any First Amendment rights, he argues that it is properly tailored to pass constitutional muster.

The evidence that the parties have proffered in this case and on which this opinion relies shows that social media has become increasingly salient in modern life. Definitions of social media vary, but generally the term refers to websites, digital applications, and the like that allow users to “self-present,” interact with one another, generate content, and consume content published by others. U.S. Surgeon General’s Advisory, Social Media and Youth Mental Health (2023), Yost Mot. Summ. J. Ex. B [hereinafter Surgeon General], R. 42-2, PageID #510. Social media uptake is pervasive. Approximately 95% of teenagers aged 13 to 17 use it. Id. at PageID #493. Data from 2021 reflected that eighth and tenth-graders used social media three and a half hours per day on average. Id. at PageID #496. One study reported “that, as of 2022, nearly half of adolescents reported being online ‘almost constantly,’ up from 24% in 2015.” Raffoul et al., Social Media Platforms Generate Billions of Dollars in Revenue from U.S. Youth: Findings from a Simulated Revenue Model, PlosOne (Dec. 27, 2023), Yost Mot. Summ. J. Ex. A [hereinafter Raffoul], R. 42-1, PageID #482.

The popularity of those platforms seems to come at a cost, as evidence has connected them to deleterious effects on children’s mental health. Id. For young people, who are at a pivotal stage in cognitive development, social media has been linked to issues with sleep, anxiety, body dysmorphia, depression, and bullying. Id.; Surgeon General, PageID #495–96. Social media affects users through not only the substantive content to which it exposes them, but also “the amount of time [they] spend on platforms, . . . the activities and interactions social media affords, and the degree to which it disrupts activities that are essential for health like sleep and physical activity.” Surgeon General, PageID #494. Features such as “[p]ush notifications, autoplay, infinite scroll, quantifying and displaying popularity (i.e., ‘likes’), and algorithms that leverage user data to serve content recommendations . . . maximize engagement.” Id. at PageID #498. Some research has suggested that social media use may have physiological effects like those from substance and gambling addictions. Id. No. 25-3371 NetChoice, LLC v. Yost Page 4

Young people are biologically and psychologically more vulnerable to those dangers than adults. See Am. Psych. Ass’n, Potential Risks of Content, Features, and Functions: A Closer Look at the Science Behind How Social Media Affects Youth 1–3 (2024). While much of the evidence currently available establishes only correlation, some experiments have signaled that social media is a driver of psychological harm to minors. See Zara Abrams, Instagram’s Effects on Mental Health, Monitor on Psychology, Mar. 2022, at 32. And while social media may “benefit[ ]some children and adolescents, there are ample indicators that social media can also have a profound risk of harm to the[ir] mental health and well-being . . . .” Surgeon General, PageID #493.

The record additionally reflects that social media imperils children by exposing them to sexual and financial predators. North Aff., Yost Mot. Summ. J. Ex. D, R. 42-4, PageID #529– 30; Surgeon General, PageID #498. Research shows that a significant number of minors using social media interact with adult strangers and experience sexual solicitation from, grooming by, and confidential relationships with those strangers. Parisa Rezaee Borj et al., Online Grooming Detection: A Comprehensive Survey of Child Exploitation in Chat Logs, Knowledge-Based Sys., Jan. 2023, Yost Mot. Summ. J. Ex. H, R. 42-8, PageID #594.

Minors’ attachment to social media is no accident. Social media operators design their sites to hold users’ attention for as long as possible. See Abrams, supra, at 30; Mark D. Griffiths, Adolescent Social Networking: How Do Social Media Operators Facilitate Habitual Use?, 36 Educ. & Health 66, 68 (2018); Julian Morgans, The Secret Ways Social Media Is Built for Addiction, Vice (May 17, 2017, 11:09 PM), https://perma.cc/W3X6-GNX3; Raffoul, R. 42-1, PageID #483. Young eyes on screens are highly profitable in terms of advertising revenue. A study from 2023 estimated that Facebook, Instagram, Snapchat, TikTok, Twitter, and YouTube, collectively, earn $11 billion in revenue from advertising directed to U.S. children between 0 and 17 years of age and $2 billion in revenue from advertising directed to those 0 to 12 years of age. Raffoul, R. 42-1, PageID #484. Across Snapchat, TikTok, and YouTube, “30– 40% of the advertising revenue” was “attributable to young people.” Id.

To capitalize on users’ engagement, social media companies leverage “personalized computational advertising to match users’ specific demographics and usage patterns with No. 25-3371 NetChoice, LLC v. Yost Page 5

advertisers’ financial interests[,]” raising “ ethical concerns regarding the amount of data . . . require[d] from vulnerable users, such as youth.” Id. at PageID #483; see also Testimony of Tim Kendall, Chief Exec. Officer, Moment, Hearing on “Mainstreaming Extremism: Social Media’s Role in Radicalizing America” Before the Subcomm. on Consumer Prot. & Com. of the H. Comm. on Energy & Com., 116th Cong. (2020) (“[W]e sought to mine as much human attention as possible and turn into historically unprecedented profits . . . [by] t[a]k[ing] a page from Big Tobacco’s playbook, working to make our offering addictive at the outset.”); Mike Allen, Sean Parker Unloads on Facebook: “God only Knows What it’s Doing to Our Children’s Brains,” Axios (Nov. 9, 2017), https://perma.cc/C4JF-FQB8 (quoting founding president of Facebook as stating that “the thought process” of the architects of social media applications was “consciously” to “consume as much of your time and conscious attention as possible” and “exploit[] a vulnerability in human psychology”).

Yost posits that social media companies’ ability to profit from minors’ engagement depends in part on those minors’ assent to certain terms and conditions precedent to creating user accounts. For example, Facebook’s Terms of Service include permissions from the user for Facebook to “to host, use, distribute, modify, run, copy, publicly perform or display, translate, and create derivative works of [the user’s] content . . . .” Terms of Service, Meta (Jan. 1, 2025), https://perma.cc/9AVV-4B4W. They also provide that Meta may “use [the user’s] name and profile picture and information about actions [the user has] taken on Facebook next to or in connection with ads, offers, and other sponsored or commercial content . . . .” Id. They disclaim all liability. Id. And they require users to submit to personal jurisdiction in the U.S. District Court for the Northern District of California and San Mateo County state court, while reserving the right for Meta to bring a claim in any competent court with jurisdiction. Id.

NetChoice “is a national trade association of online businesses that share the goal of promoting free speech and free enterprise on the Internet.” Szabo Decl. Supp. Pl.’s Mot. TRO & Prelim. J., R. 2-1, PageID #64. Its members (the Members) comprise

a broad array of popular websites, including: Airbnb, Alibaba.com, Amazon.com, AOL, Dreamwidth, eBay, Etsy, Expedia, Fluid Truck, Google, HomeAway, Hotels.com, Lime, Lyft, Meta, Nextdoor, Oath, OfferUp, Orbitz, PayPal, Pindrop, Pinterest, Snap Inc., StubHub, Swimply, TikTok, TravelTech, Travelocity, No. 25-3371 NetChoice, LLC v. Yost Page 6

Trivago, Turo, Verisign, VRBO, VSBLTY, Waymo, Wing, X (formerly known as Twitter), and Yahoo!.

Id. According to NetChoice, some Members’ sites are considered social media sites and others are not. Many Members “have developed content-moderation policies and other safeguards to protect minors online[,]” such as limiting account creation to users 13 years of age or older, moderating the content that minors encounter on the platforms, limiting user speech that Members consider harmful, and offering tools to parents and guardians to control what their children can access. Id. at PageID #66.

On July 4, 2023, the governor of Ohio signed the Act into law. Section (A)(1) defines an Operator as:

any business, entity, or person that operates an online web site, service, or product that has users in this state and that allows those users to do all of the following: (a) Interact socially with other users within the confines of the online web site, service, or product; (b) Construct a public or semipublic profile for the purpose of signing into and using the online web site, service, or product; (c) Populate a list of other users with whom an individual shares or has the ability to share a social connection within the online web site, service, or product; (d) Create or post content viewable by others, including on message boards, chat rooms, video channels, direct or private messages or chats, and a landing page or main feed that presents the user with content generated by other users.

Parental Notification by Social Media Operators Act (the Act) § (A)(1), H.B. 33, 135th Gen. Assemb., Reg. Sess. (Ohio 2023) (codified at Ohio Rev. Code § 1349.09(A)(1)). Basically, the Act covers any site that allows users in Ohio to interact socially with one another, build profiles, accumulate social connections, and create or post content that one another can view. Section (B) of the Act institutes requirements for any such “operator of an online web site, service, or product that targets [] or is reasonably anticipated to be accessed by [unemancipated] children” under the age of 16 (Children),1 (a Covered Operator). Ohio Rev. Code §§ 1349.09(A)(2), (B). Section (C) of the Act provides an optional list of 11 factors to assist in the inquiry of whether a site targets or is reasonably anticipated to be accessed by Children, namely:

1 Where this opinion refers to “Children,” capitalized, it refers to Children as defined by the Act: unemancipated children under the age of 16. Where it uses the lower-case term “children,” it refers to children as commonly understood. That guideline does not apply to quoted material, which reflects the capitalization that appears in the relevant source. No. 25-3371 NetChoice, LLC v. Yost Page 7

(1) Subject matter; (2) Language; (3) Design elements; (4) Visual content; (5) Use of animated characters or child-oriented activities and incentives; (6) Music or other audio content; (7) Age of models; (8) Presence of child celebrities or celebrities who appeal to children; (9) Advertisements; (10) Empirical evidence regarding audience composition; and (11) Evidence regarding the intended audience.

Id. § 1349.09(C).

Sections (N) and (O) then exempt from the Act’s coverage any site for which (1) “the predominant or exclusive function is: (a) Cloud storage or cloud computing services; (b) Broadband internet access services; [or] (c) Search engine services”, id. § 1349.09(N); or (2) “interaction between users is limited to” either reviewing and commenting on reviews of commercial products or “[c]omments incidental to content posted by an established and widely recognized media outlet, the primary purpose of which is to report news and current events[,]” id. § 1349.09(O).

The Act requires a Covered Operator to: (1) “[o]btain verifiable consent for any contract with a child . . . from the child’s parent or legal guardian using any of [several listed] methods”; (2) “[p]resent to the child’s parent or legal guardian a list of the [Operator’s site’s] features . . . related to censoring or moderating content”; (3) provide the parent or guardian a link to review those features at a later time; (4) confirm or verify the parental consent by written mail or telephone; and (5) in the absence of parental consent, “deny the child access to or use of the [site.]” Id. §§ 1349.09(B), (D), (E).

The Act provides that “[t]he attorney general shall investigate any noncompliance with this section in the same manner . . . as in section 1349.191 of the Revised Code.” Id. § 1349.09(G). Section 1349.191, in turn, provides, “The attorney general may conduct an investigation if the attorney general, based on complaints or the attorney general’s own inquiries, has reason to believe” an entity is noncompliant. Id. § 1349.191(B). In that investigation, “the attorney general may administer oaths, subpoena witnesses, adduce evidence, and subpoena the production of any book, document, record, or other relevant matter.” Id. § 1349.191(C). If the subject of such a subpoena fails to cooperate, the court of common pleas may issue orders compelling compliance with the subpoena, adjudging the subpoenaed entity in contempt of court, No. 25-3371 NetChoice, LLC v. Yost Page 8

or granting injunctive or other relief. Id. § 1349.191(F). The court additionally may impose a civil penalty on an entity that violates such an order. Id. § 1349.191(G).

“If it appears that an operator . . . failed to comply with [the Act], the attorney general has the exclusive authority to bring a civil action . . . for appropriate relief including a temporary restraining order, preliminary or permanent injunction, and civil penalties.” Id. § 1349.09(H). Then, “[i]f a court finds that an operator . . . entered into a contract with a child without consent of the child’s parent or guardian, as required by this section, the court shall impose a civil penalty” of up to $1,000 for each of the first 60 days of noncompliance, up to $5,000 for each subsequent day of noncompliance up to the 90th day, and up to $10,000 for each subsequent day of noncompliance after that. Id. § 1349.09(I). A noncompliant Covered Operator is additionally liable for the Attorney General’s investigation and litigation costs. Id. § 1349.09(K).

The Attorney General must give notice before commencing a civil action only if the subject Covered Operator is in “substantial compliance[,]” in which case the noticed Covered Operator has 90 days to cure the violation and document the measures it has taken to prevent future violations. Id. § 1349.09(M).

The Act took effect on January 15, 2024. Am. Sub. H.B. 33 § 803.380, 135th Gen. Assemb., Reg. Sess., 2023 Ohio Laws File 8.

On January 5, 2024, NetChoice filed a complaint in the U.S. District Court for the Southern District of Ohio, under 42 U.S.C. § 1983 and 28 U.S.C. § 2201, suing Defendant Yost in his official capacity as the Attorney General of Ohio. NetChoice claimed that the Act was unconstitutional because it “imposes blanket parental-consent requirements for minors to access and engage in all manner of protected speech across a wide swath of websites[,]” might deter potential users from creating accounts, discriminates among Operators based on the content of their speech, and is unconstitutionally vague. Compl., R. 1, PageID #3. NetChoice sought declaratory relief, an injunction prohibiting enforcement of the Act against NetChoice and its Members, and attorneys’ fees and costs.

On the same day that NetChoice initiated the lawsuit, it moved the district court for a temporary restraining order and preliminary injunction prohibiting Yost from enforcing the Act No. 25-3371 NetChoice, LLC v. Yost Page 9

against NetChoice’s Members. The district court granted both, finding NetChoice likely to succeed on its First Amendment and vagueness claims. NetChoice, LLC v. Yost, 711 F. Supp. 3d 844, 856–59 (S.D. Ohio 2024); NetChoice, LLC v. Yost, 716 F. Supp. 3d 539, 560–62 (S.D. Ohio 2024). The parties agreed to forego discovery and cross-moved for summary judgment.

On April 16, 2025, the district court entered judgment in favor of NetChoice and enjoined enforcement of the Act entirely. NetChoice, LLC v. Yost, 778 F. Supp. 3d 923, 959 (S.D. Ohio 2025). It held that NetChoice had prudential standing to assert the rights of Children in Ohio who use or might use Member sites (Children Users). Id. at 946. The court decided that any conflicts between NetChoice and Children Users did not pertain to “the right of minors to access social media platforms without obtaining parental consent.” Id. at 944. The district court additionally held that the Act was facially unconstitutional because “in every application to a covered website, the Act raises the same First Amendment issues.” Id. at 947. The Act, the district court concluded, implicated protected speech, id. at 950, was content-based, id. at 953– 54, required parental consent for Children to access constitutionally protected and non-obscene content, id.at 954–55, failed strict scrutiny, id. at 957, and was unconstitutionally vague, id. at 957–58. Therefore, NetChoice succeeded on the merits. Finding the other requirements for a permanent injunction satisfied, the district court awarded NetChoice that relief. Id. at 959.

Yost appeals from that order and judgment.

II. DISCUSSION

Yost raises four issues on appeal. He contends that (1) NetChoice lacks prudential standing to assert the rights of Members’ Children Users; (2) the district court should not have granted summary judgment in favor of NetChoice on its First Amendment facial challenge; (3) the district court should not have granted summary judgment in favor of NetChoice on its vagueness challenge; and (4) the district court’s relief was overly broad because it invalidated the entire law and enjoined enforcement universally. NetChoice disagrees in all respects, except that it does not meaningfully contest that an injunction against enforcement in this case should apply only to the covered Members. No. 25-3371 NetChoice, LLC v. Yost Page 10

A. Standing

The parties do not dispute that NetChoice has associational standing to bring this suit on behalf of its Members, but rather whether NetChoice may assert the rights of third-party Children Users. According to Yost, NetChoice does not meet the prudential requirements to represent the First Amendment rights of Children Users because of significant conflicts of interest between them. NetChoice responds that it does have standing to assert those rights because enforcement of the Act against Members would violate Children Users’ rights, third- party standing is permissive in the First Amendment context, and NetChoice’s and Children Users’ interests are aligned with respect to the issues in this case.

i. Standard of Review

This Court reviews a decision to grant summary judgment on the issue of standing de novo. Fenner v. Gen. Motors, LLC, 113 F.4th 585, 593 (6th Cir. 2024) (citing B & H Med., L.L.C. v. ABP Admin., Inc., 526 F.3d 257, 264 (6th Cir. 2008)).

ii. Analysis

Under Article III of the U.S. Constitution, federal courts’ jurisdiction is “limit[ed] . . . to ‘Cases’ and ‘Controversies.’” Susan B. Anthony List v. Driehaus, 573 U.S. 149, 157 (2014) (quoting U.S. Const. art. III, § 2). Standing doctrine derives from that limitation and reserves the judicial process for disputes among litigants as opposed to the kind of lawmaking attendant to the other branches of government. See id. Article III standing requires a plaintiff to present “(1) an injury in fact that is both (2) caused by the defendant’s conduct and (3) redressable by a favorable court decision.” Christian Healthcare Ctrs., Inc. v. Nessel, 117 F.4th 826, 842 (6th Cir. 2024) (citing Driehaus, 573 U.S. at 157–58). A party seeking summary judgment on the issue of standing must eliminate any genuine dispute of material fact regarding each element of standing. Kareem v. Cuyahoga Cnty. Bd. of Elections, 95 F.4th 1019, 1022 (6th Cir. 2024) (quoting Lujan v. Defs. of Wildlife, 504 U.S. 555, 561 (1992)). On appeal, the parties do not dispute whether NetChoice has the “irreducible constitutional minimum of standing” to bring this action on behalf of its Members, based on a theory of associational standing. Lujan, 504 U.S. at 560; see Ass’n of Am. Physicians & Surgeons v. U.S. Food & Drug Admin., 13 F.4th 531, No. 25-3371 NetChoice, LLC v. Yost Page 11

537 (6th Cir. 2021) (stating the requirements for an organization to sue on behalf of its members). Therefore, we do not evaluate that theory in this case.

In addition to the Article III requirements, courts have developed prudential limitations on standing, “as a tool of ‘judicial self-governance.’” See Prime Media, Inc. v. City of Brentwood, 485 F.3d 343, 349 (6th Cir. 2007) (quoting Warth v. Seldin, 422 U.S. 490, 500 (1975)). Among them is the general rule “preclud[ing] litigation in federal court . . . where instead of litigating ‘his own legal rights and interests,’ the plaintiff instead purports to ‘rest his claim to relief on the legal rights or interests of third parties.’” Id. (quoting Warth, 422 U.S. at 499). Generally, parties may not assert the rights of nonparties. Connection Distrib. Co. v. Reno, 154 F.3d 281, 295 (6th Cir. 1998). That principle is important both because it “frees the Court . . . from unnecessary pronouncement on constitutional issues [and] . . . premature interpretations of statutes in areas where their constitutional application might be cloudy” and “assures the court that the issues before it will be concrete and sharply presented.” Sec’y of State of Md. v. Joseph H. Munson Co., 467 U.S. 947, 955 (1984) (first quoting United States v. Raines, 362 U.S. 17, 22 (1960); and then citing Baker v. Carr, 369 U.S. 186, 204 (1962)). Moreover, the third-party rightsholders might have considered reasons for choosing not to assert their rights in court. See Singleton v. Wulff, 428 U.S. 106, 113–14 (1976) (plurality opinion) (“[I]t may be that in fact the holders of those rights [] do not wish to assert them . . . .”). For those reasons and because of the importance of “effective advocacy” for arriving at correct conclusions, courts should “construe legal rights only when the most effective advocates of those rights are before them.” Id. at 114; see also Kowalski v. Tesmer, 543 U.S. 125, 129 (2004) (“This rule assumes that the party with the right has the appropriate incentive to challenge (or not challenge) governmental action and to do so with the necessary zeal and appropriate presentation.” (citing Warth, 422 U.S. at 500)).

But courts have made exceptions to that prudential rule. Connection Distrib., 154 F.3d at 295 (citing Edmonson v. Leesville Concrete Co., 500 U.S. 614, 629 (1991)). Third-party standing is one of them. “A litigant may raise a claim on behalf of a third party if the litigant can demonstrate that it has suffered a concrete, redressable injury, that it has a close relation with the third party, and that there exists some hindrance to the third party’s ability to protect his or her No. 25-3371 NetChoice, LLC v. Yost Page 12

own interests.” Id. (citing Edmonson, 500 U.S. at 629). The litigant must have a relationship to the rightsholder “such that the litigant is an effective proponent of the rights of the third party . . . .” Id. (citing Singleton, 428 U.S. at 114–16 (plurality opinion)).

Yost argues that NetChoice and Children Users do not share a sufficiently close relationship to warrant third-party standing because of a pertinent conflict of interest. NetChoice cannot effectively advocate on behalf of Children Users, he submits, because NetChoice and its Members are interested in exploiting minors’ engagement with Members’ sites regardless of what outcome is in the Children Users’ best interests. Yost is persuasive on that point. NetChoice has certainly argued for unfettered access to social media for minors. But the Act intends to mitigate harm to minors associated with increased use of social media and the lopsided contracts that NetChoice has minors execute before they may enjoy that access. Without Children Users present in the litigation, we cannot be sure that a well-informed, rational Child would reject the Act’s protections or advocate for precisely the same contours of First Amendment doctrine as NetChoice does.

NetChoice leans on Virginia v. American Booksellers Association, 484 U.S. 383 (1988). NetChoice is correct that a litigant may bring a First Amendment overbreadth challenge that is based not on the litigant’s own injury but on “a judicial prediction or assumption that the statute’s very existence may cause others not before the court to refrain from constitutionally protected speech or expression.” Am. Booksellers, 484 U.S. at 392–93 (quoting Munson, 467 U.S. at 956–957). Given the unique concerns about freedom of expression that underlie the First Amendment, the Supreme Court has loosened the requirements for prudential standing in overbreadth challenges to allow a litigant to challenge the facial validity of a statute that might be applied constitutionally to the litigant. Broadrick v. Oklahoma, 413 U.S. 601, 612 (1973) (quoting Dombrowski v. Pfister, 380 U.S. 479, 486 (1965)). Such leniency in this setting is grounded in the idea that the harm of muting any protected speech outweighs the risk of permitting unprotected speech to proceed unrestrained. Id. The concern is the “possibility that, rather than risk punishment for [] conduct in challenging the statute, [a rightsholder] will refrain from engaging further in the protected activity” and therefore “[s]ociety as a whole [] would be the loser.” Munson, 467 U.S. at 956. No. 25-3371 NetChoice, LLC v. Yost Page 13

But the case law does not establish a simplistic rule that wherever First Amendment overbreadth questions are concerned, they overshadow all other prudential considerations in the third-party standing analysis. American Booksellers acknowledged unique concerns that arise in a First Amendment challenge and gave courts leeway to factor those concerns into the balance, but the background motivations of the prudential standing doctrine do not evaporate because the claim derives from the First Amendment.

Our sister circuit’s analysis in Amato v. Wilentz, 952 F.2d 742 (3d Cir. 1991), is helpful. In that case, the New Jersey Supreme Court’s Chief Justice blocked Warner Brothers from using state courthouses as a film set. 952 F.2d at 742. In exchange for permission to film, Warner Brothers had offered a $250,000 donation to the Courthouse Restoration Fund. Id. Essex County, home to one of the courthouses, sued the Chief Justice, claiming that he had violated Warner Brothers’ First Amendment rights. Id. The court of appeals held that Essex County lacked third-party standing to assert Warner Brothers’ First Amendment rights. Id. at 744.

The court reasoned that the Supreme Court precedent on third-party standing had “(1) identif[ied] factors that are relevant to determining third party standing and (2) render[ed] an overall balance of factors dispositive.” Id. at 750. Those cases had not dictated bright-line, obligatory rules. Thus even where the relationship between the litigant and third party is “analogous” to the relationships in cases where the Supreme Court has found third-party standing, the “exception . . . should not apply [if] its assumptions do not hold.” Id. at 752. In Amato, the “critical assumption . . . —namely, an identity of interests—[did] not obtain.” Id. Warner Brothers might have decided against bringing legal action out of concern that the suit would fail and “create[e] an undesirable legal precedent[] or backfire in practice.” Id. at 753. “[A]lthough the County’s advocacy of its own interests ha[d] been quite competent and extremely vigorous, [the court] fear[ed] that the County may not have adequately advocated Warner Brothers interests” and thus was “disinclined to allow third party standing . . . .” Id. Amato did not confront an overbreadth challenge specifically, and the court merely speculated that “if Essex . . . had brought a challenge to the . . . system, the overbreadth cases might well counsel in favor of according third party standing.” Id. at 754. But that dictum also suggests No. 25-3371 NetChoice, LLC v. Yost Page 14

that the cases might not so counsel. In some instances, an overbreadth claim might overcome the lack of an “identity of interests,” but not always.

In this case, it cannot be predicted or assumed that the Act will chill constitutionally protected speech among minors in the manner contemplated in American Booksellers. In that case, booksellers challenged a statute that outlawed the commercial display of “visual or written material that ‘depicts sexually explicit nudity, sexual conduct or sadomasochistic abuse and which is harmful to juveniles’” where “juveniles [might] examine and peruse” it. Am. Booksellers, 484 U.S. at 386 (quoting Va. Code Ann. § 18.2–391(A)). The Court’s worry in that case was that the statute would chill protected conduct because adult book buyers “would be reluctant to enter an ‘adults only’ store or section of a store” as would be required under the statute. Id. at 389; see also Bates v. State Bar of Arizona, 433 U.S. 350, 380 (1977) (“The reason for the special rule . . . is [that] . . . First Amendment interests are fragile interests, and a person who contemplates protected activity might be discouraged by the in terrorem effect of the statute.”). In a sense, the hurdle of obtaining parental consent or the prospect of enhanced parental oversight might chill Children Users’ speech on social media. But that chill would appear to be different in kind. Minors are unlikely to be chilled from speaking their minds due to stigma or because the Act reflects a societal distaste for certain ideas. The Act therefore does not raise meaningful concerns about muting valuable protected discourse.

That distinction weakens the call to set aside our prudential rules. As in Amato, even though the plaintiff seeks to assert the First Amendment rights of a non-party at the opposite end of a speech-related transaction, the lack of “an identity of interests” carries more weight. 952 F.2d at 752. We should not ignore the possibility that Children Users might independently decide against challenging the Act out of caution against generating unfavorable caselaw or perpetuating harm to themselves. NetChoice’s most vigorous advocacy for expansive First Amendment doctrine is not necessarily the most effective advocacy for Children Users’ interests. Even if a maximalist construction of the First Amendment right were entirely exploitative of and deleterious to minors, NetChoice would presumably still support it. Therefore, permitting NetChoice to challenge the facial validity of the Act based on Children Users’ First Amendment rights would be imprudent. No. 25-3371 NetChoice, LLC v. Yost Page 15

I read Kowalski v. Tesmer, 543 U.S. 125 (2004), differently from Judge Batchelder and do not think that it negates the applicability of Amato to this case. First, Kowalski was not a First Amendment case, and its characterization of that area of law is dicta. Amato, meanwhile, was a First Amendment case. Second, Kowalski’s reference to third-party standing in First Amendment cases explains that the close relationship and hindrance requirements have been lessened in that context. 543 U.S. at 130. That statement is compatible with the use of a more fluid balancing approach in First Amendment cases. Kowalski’s reasoning would only render the burden higher for a party to establish third-party standing, so it would not affect the conclusion that under a looser standard, NetChoice loses.

Referencing cases in which courts have permitted vendors to assert the rights of their customers, NetChoice argues that not all of a litigant’s interests must be aligned with the third party’s interests before the litigant may enjoy third-party standing. It says that the conflict analysis should “focus[] on the nature of the claim asserted.” NetChoice Br. 28 (citation omitted). That argument fails. First, while not any conflict of interest abrogates third-party standing, the unique conflict in this case connects distinctly to the right at issue. The claims and defenses in this case inextricably intertwine NetChoice’s financial and legal incentives with the significant potential harm to the Children rightsholders, all centered around the core third-party First Amendment right. The special profitability of minors’ social media habits is, conceivably, the cause of both the danger that Ohio has identified and NetChoice’s purported interest in the Children Users’ First Amendment rights. Second, the relationship between NetChoice and Children Users is not like typical vendor-customer relationships. The customer is not walking into a vendor’s shop to execute a one-off transaction. Yost has offered evidence that the transactions at issue involve lopsided and exploitative contracts that initiate enduring relationships in which Members seek to capture and hold Children Users’ attention for as long as possible, even if that engagement is harmful.

To be clear, the dynamic before us, wherein the very issues in the case create serious tension between NetChoice and the Children Users’ interests, is sui generis. Hewing closely to the facts at hand, I do not expect that future cases in which speaker and platform are better aligned will result in the same outcome. The premise of this case distinguishes it from others No. 25-3371 NetChoice, LLC v. Yost Page 16

where the dispute does not so credibly and directly pit the plaintiff’s self-interest against the third-party’s wellbeing. I do not imply anything, for example, about a doctor’s competence to assert a patient’s rights, or a school’s ability to assert a student’s. But on these facts, NetChoice is not the proper party to litigate the rights of Children Users. I would decline to recognize third- party standing for NetChoice to rest its overbreadth challenge on allegations that the Act violates Children Users’ First Amendment rights.

B. NetChoice’s First Amendment Challenge

Before us is the district court’s order granting summary judgment to NetChoice on its facial challenge to the Act. NetChoice, 778 F. Supp. at 931, 936. NetChoice brought both a traditional facial challenge and, in the alternative, a First Amendment overbreadth challenge. Id. at 936. To prevail on a traditional facial challenge, a plaintiff must show that the law has no constitutional applications whatsoever. Moody v. NetChoice, LLC, 603 U.S. 707, 723 (2024). For a First Amendment overbreadth challenge, courts have lightened the load, allowing a plaintiff to prevail “if a substantial number of [the law’s] applications are unconstitutional, judged in relation to the statute’s plainly legitimate sweep.” Ams. for Prosperity Found. v. Bonta, 594 U.S. 595, 615 (2021) (quoting United States v. Stevens, 559 U.S. 460, 473 (2010)). “Even in First Amendment cases, facial invalidation is ‘strong medicine that is not to be casually employed.’” Schickel v. Dilger, 925 F.3d 858, 880 (6th Cir. 2019) (quoting Connection Distrib. Co. v. Holder, 557 F.3d 321, 336 (6th Cir. 2009) (en banc)); see also Broadrick, 413 U.S. at 613 (“The consequence . . . is that any enforcement of a statute thus placed at issue is totally forbidden [absent] a limiting construction or partial invalidation . . . . [T]he overbreadth doctrine . . . has been employed by the Court sparingly and only as a last resort.”).

NetChoice argues, and the district court found, that the First Amendment protects the speech exchanged on Member sites and that the Act is facially unconstitutional because it burdens that speech with a content-based parental-consent requirement that fails strict scrutiny. NetChoice, 778 F. Supp. 3d at 950, 953–55, 957. Yost argues that NetChoice has not established that the Act is unconstitutional, let alone that it is facially unconstitutional. He submits that the Act regulates non-expressive conduct, triggering only rational basis review, that even if No. 25-3371 NetChoice, LLC v. Yost Page 17

heightened scrutiny applies, the Act passes, and that even if it did not, NetChoice has not shown that the Act’s unconstitutional applications substantially outweigh the constitutional ones.

This Court reviews a grant of summary judgment de novo. Schickel, 925 F.3d at 869 (citing Allied Constr. Indus. v. City of Cincinnati, 879 F.3d 215, 219 (6th Cir. 2018)). Summary judgment is proper “only if ‘there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.’” Id. (quoting Fed. R. Civ. P. 56). In the event of cross- motions for summary judgment, the Court “draw[s] all reasonable inferences against the party whose motion is under consideration.” Id. (quoting McKay v. Federspiel, 823 F.3d 862, 866 (6th Cir. 2016)).

The Act Burdens Protected Speech

The First Amendment to the U.S. Constitution states, “Congress shall make no law . . . abridging the freedom of speech, or of the press . . . .” U.S. Const. amend I. The Fourteenth Amendment incorporated that prohibition against the states. Manhattan Cmty. Access Corp. v. Halleck, 587 U.S. 802, 808 (2019).

The U.S. Supreme Court has recognized that, at least to some degree, social media encompasses protected speech. See Packingham v. North Carolina, 582 U.S. 98, 102, 108 (2017) (holding that a law prohibiting a person convicted of a sex offense from accessing Facebook was unconstitutional); Moody, 603 U.S. at 716 (“To the extent that social-media platforms create expressive products, they receive the First Amendment’s protection.”). Social media platforms, “in at least some functions, are indeed engaged in expression” such as when they “make choices about what third-party speech to display and how to display it.” Moody, 603 U.S. at 716. That role is akin to that of “[t]raditional publishers and editors . . . .” Id. at 717. But not all of a social media platform’s conduct is necessarily expressive; a platform might merely provide a forum for the public to exchange speech. See id. at 730–31 (analogizing to a shopping mall that allows the public to distribute handbills on its property and a school hosting No. 25-3371 NetChoice, LLC v. Yost Page 18

employer interviews). To the extent that a law restricts the expressive functions of social media, though, it burdens First Amendment-protected speech. See TikTok Inc. v. Garland, 604 U.S. 56, 69 (2025) (per curiam) (“[A]n effective ban on a social media platform with 170 million U.S. users certainly burdens those users’ expressive activity in a non-trivial way.”).

NetChoice’s theory is that Covered Operators engage in their own protected expression by curating third-party speech. In fact, the curatorial tools Operators use in that regard, such as data-based algorithms, are precisely some of the features that seem to be harmful to children. The selection of content for publication to a User’s feed is the kind of editorial decision-making that the First Amendment protects.

A Covered Operator may not engage in expression if it, for example, provides a forum for Users to communicate but does not choose what material to publish or how to display it. Co- owner of NetChoice Member Dreamwidth Studios, LLC (Dreamwidth), for example, stated that Dreamwidth “do[es] not offer any algorithmic sorting or display of user timelines that adjusts the display of content based on a prediction that a particular user will be more or less interested in a particular piece of content . . . .” Paolucci Decl. Supp. Pl.’s Mot. TRO & Prelim. Inj., R. 2-2, PageID #76. But Yost does not seriously contest that the Members all curate content to some degree, so I accept that assumption for the purposes of this opinion.

To the extent that NetChoice’s Members engage in protected expression, the Act burdens that protected speech. The Act imposes a parental-consent requirement that would inhibit Covered Operators’ preferred method of disseminating its content. Beyond imposing rules that Covered Operators must follow, the Act would also conceivably affect the magnitude of the audience that Members’ speech reaches. Some Children will not be able to obtain parental consent, and some Children who could obtain it may yield to the additional effort of doing so. The Act additionally risks chilling Members’ engagement in speech that Children might reasonably be likely to access, or even speech outside of that category, out of caution. Therefore, the Act burdens at least some protected speech. No. 25-3371 NetChoice, LLC v. Yost Page 19

Strict Scrutiny Applies

While “legislatures and agencies . . . will generally be better positioned than courts to respond to the emerging challenges social-media entities pose[,]” courts must safeguard the expression that occurs on social media. Moody, 603 U.S. at 716. The government may restrict protected activities under some circumstances. See Heffron v. Int’l Soc. for Krishna Consciousness, Inc., 452 U.S. 640, 647 (1981) (“[T]he First Amendment does not guarantee the right to communicate one’s views at all times and places or in any manner that may be desired.” (citing cases)). But courts scrutinize “statutes which, although directed at activity with no expressive component, impose a disproportionate burden upon those engaged in protected First Amendment activities.” TikTok, 604 U.S. at 67–68 (quoting Arcara v. Cloud Books, Inc., 478 U.S. 697, 703–704 (1986)).

Yost argues that the Court should apply rational basis review because the Act does not regulate speech. But even if the Act directly regulated only the non-expressive conduct of contracting, the Act would still burden at least some protected speech. Rational basis review is reserved for “laws that do not implicate ‘fundamental constitutional rights’ at all.” Free Speech Coal., Inc. v. Paxton, 606 U.S. 461, 495 (2025) (quoting F.C.C. v. Beach Commc’ns, Inc., 508 U.S. 307, 313 (1993)). Thus, even unintentional effects on speech may require scrutiny more rigorous than rational basis review.

The level of heightened scrutiny that attaches to a law depends on whether the law is content-based. Planet Aid v. City of St. Johns, 782 F.3d 318, 326 (6th Cir. 2015). Content neutrality means neutrality not only as to viewpoint but also as to topic or subject matter. Id. at 326–27 (quoting Consol. Edison Co. of N.Y. v. Pub. Serv. Comm’n of N.Y., 447 U.S. 530, 536 (1980)). The government’s motivation behind a law is the key determinant of content neutrality. Connection Distrib., 154 F.3d at 290 (quoting Ward v. Rock Against Racism, 491 U.S. 781, 791 (1989)). The government’s stated purpose, however, is not the be-all and end-all. A law receives strict scrutiny if it facially discriminates based on content, regardless of whether the legislature had an “[i]llicit legislative intent.” Reed v. Town of Gilbert, 576 U.S. 155, 165 (2015) (quoting Simon & Schuster, Inc. v. Members of N.Y. State Crime Victims Bd., 502 U.S. 105, 117 (1991)); see also Turner Broad. Sys., Inc. v. F.C.C., 512 U.S. 622, 642 (1994) (“[W]hile a No. 25-3371 NetChoice, LLC v. Yost Page 20

content-based purpose may be sufficient in certain circumstances to show that a regulation is content based, it is not necessary to such a showing in all cases.”).

Determining whether a law is content-based is nuanced, but generally a law that “focuses on the noncommunicative aspects of the speech[] and treats speech the same regardless of what the speech says[ is] content-neutral.” Planet Aid, 782 F.3d at 326 (quoting Eugene Volokh, The First Amendment and Related Statutes 360 (5th ed. 2014)). A few considerations are: (1) whether the government enacted the law due to disagreement with the message; (2) whether the regulation affects the speaker’s “communicative impact”; (3) whether the “predominant intent” pertained to the content or secondary effects of the speech; and (4) whether the law “is ‘based on the content of speech’” or is “applicable . . . irrespective of content . . . .” Id. at 326–27 (citing cases). Exceptions in a law “can render [it] content-based to the extent [they] require[] consideration of the topic or subject matter of the speech to determine whether the speech qualifies . . . .” Zillow, Inc. v. Miller, 126 F.4th 445, 462 (6th Cir. 2025) (citing Norton Outdoor Advert., Inc. v. Village of St. Bernard, 99 F.4th 840, 850 (6th Cir. 2024)).

The Act’s parental-consent requirement discriminates based on subject matter. While Ohio may not have intended to single out any topic or message, the effect of the Act on speakers’ communicative impact facially varies based on the content of the speech. The parental-consent requirement applies only to Covered Operators, Ohio Rev. Code § 1349.09(B), and the provisions defining that group are content-based in two ways. First, the coverage definition advises the Attorney General and courts to consider factors that are undeniably content-focused to determine whether an Operator’s site “is reasonably anticipated to be accessed by children,” such as “[s]ubject matter;” “[v]isual content;” “[m]usic or other audio content;” and the “[p]resence of . . . celebrities who appeal to children . . . .” Id. § 1349.09(C). The Act thus disfavors subject matter that appeals to Children. Second, the Act excludes from coverage Operators whose Users interact only in “[c]omments incidental to content posted by an established and widely recognized media outlet, the primary purpose of which is to report news and current events.” Id. §1349.09(O). The Act thus favors news and current events subject matter. Assessing the applicability of that exception to an Operator requires consideration of the subject matter of the Operator’s speech. Neither of those provisions focuses purely on No. 25-3371 NetChoice, LLC v. Yost Page 21

noncommunicative aspects of speech. Although the state’s purpose was not necessarily to disagree with a particular message, but rather to target the secondary effects of Operators’ speech, the Act applies differently based on the substance of that speech.

Generally, the government may not “restrict expression” based on content, aside from limited exceptions including obscenity, incitement, and fighting words. Brown v. Ent. Merchants Ass’n, 564 U.S. 786, 790–91 (2011) (first quoting Ashcroft v. Am. C.L. Union, 535 U.S. 564, 573 (2002); and then citing cases). Otherwise, such restrictions receive strict scrutiny. Reed, 576 U.S. at 166. The Act is content-based and does not fit within those limited recognized exceptions. Therefore, strict scrutiny applies.

The Act Survives Strict Scrutiny

Under strict scrutiny, Yost must “prove that the restriction furthers a compelling interest and is narrowly tailored to achieve that interest . . . .” Reed, 576 U.S. at 171 (quoting Ariz. Free Enter. Club’s Freedom Club PAC v. Bennett, 564 U.S. 721, 734 (2011)). Narrow tailoring for strict scrutiny means that the law must be “the least restrictive means” of achieving the objective. Free Speech Coal., 606 U.S. at 471 (quoting McCullen v. Coakley, 573 U.S. 464, 478 (2014)). That standard is “demanding[,]” and a law will rarely satisfy it. Brown, 564 U.S. at 799. But it is not “fatal in fact.” Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 237 (1995) (quoting Fullilove v. Klutznick, 448 U.S. 448, 519 (1980) (Marshall, J., concurring in judgment)); see Williams-Yulee v. Fla. Bar, 575 U.S. 433, 437, 444 (2015) (holding that state laws prohibiting judicial candidates from personally soliciting campaign funds survived First Amendment strict scrutiny); see also Reform Am. v. City of Detroit, 37 F.4th 1138, 1143, 1155–57 (6th Cir. 2022) (holding that division of protesting groups based on their viewpoints survived strict scrutiny); KenAmerican Res., Inc. v. United States Sec’y of Lab., 33 F.4th 884, 887, 893 (6th Cir. 2022) (holding that a provision of the Federal Mine Safety and Health Act of 1977 satisfied strict scrutiny); Platt v. Bd. of Comm’rs on Grievances & Discipline of Ohio Sup. Ct., 894 F.3d 235, 241, 261–64 (6th Cir. 2018) (holding that Ohio’s fundraising and advocacy limitations on judicial candidates satisfied strict scrutiny); Grider v. Abramson, 180 F.3d 739, 742, 749, 751, 753 (6th Cir. 1999) (holding that the content-based elements of an emergency crowd control plan satisfied strict scrutiny). No. 25-3371 NetChoice, LLC v. Yost Page 22

Yost asserts that Ohio’s interests underpinning the Act are threefold: protecting children, involving parents, and regulating contracting with children. The compelling state interest in protecting children’s physical and psychological welfare is well established. See Sable Commc’ns of California, Inc. v. F.C.C., 492 U.S. 115, 126 (1989) (“We have recognized that there is a compelling interest in protecting the physical and psychological well-being of minors.”) That interest is enough, and due to the nature of the harms at issue, the other two asserted interests are in service of the first.

NetChoice contends that the state’s power to protect children from harm does not bestow on the state “a free-floating power to restrict the ideas to which children may be exposed” and that the causal link between social media use and Yost’s alleged harms is unsupported by evidence as required under Brown v. Entertainment Merchants Association, 564 U.S. 786 (2011). NetChoice Br. 53 (quoting 564 U.S. at 794). However, this case differs from Brown.

First, Ohio does not restrict what ideas Operators may circulate to Children. Rather the Act mitigates the “objective effects” of social media use. Brown, 564 U.S. at 799. In Brown, the legislation facially blocked minors from purchasing or renting violent video game content. Id. at 789. The Act, by contrast, says nothing about what subject matter Children Users may access.

Second, the state’s supporting evidence in Brown was weaker than Yost’s evidence here. In Brown, California “acknowledge[d] that it [could ]not show a direct causal link between violent video games and harm to minors[,]” relied on studies that had “been rejected by every court to consider them” and most of which “suffer[ed] from significant, admitted flaws in methodology[,]” and “show[ed] at best some correlation [with] minuscule real-world effects, such as children’s feeling more aggressive or making louder noises in the few minutes after playing a violent game . . . .” Id. at 799–800. Yost relies on findings by the Surgeon General and researchers who have identified a potentially significant connection between social media and serious harms such as depression and body dysmorphia. Additionally, the evidence in Brown suggested that the effects of violent video games were materially indistinguishable from the effects from other sources of violent material. Id. at 800–01. NetChoice has not brought to our attention evidence that other forms of media have the same negative effects on minors as social media. The link between social media and minors’ wellbeing seems “neither novel nor No. 25-3371 NetChoice, LLC v. Yost Page 23

implausible[,]” and therefore Ohio has met the “quantum of empirical evidence needed to satisfy heightened judicial scrutiny of [its] legislative judgment[] . . . .” OPAWL - Bldg. AAPI Feminist Leadership v. Yost, 118 F.4th 770, 780 (6th Cir. 2024).

Furthermore, attacks on the strength of Yost’s evidence at most create a dispute of material fact, which would weigh against summary judgment. See Turner Broad., 512 U.S. at 664–65, 668 (plurality opinion) (vacating summary judgment because “genuine issues of material fact” remained, including “whether the Government ha[d] adequately shown that the economic health of local broadcasting [was] in genuine jeopardy and in need of the protections afforded”).

NetChoice additionally argues that Ohio has not distinguished its interest in parental oversight from the interest in aiding parental authority that Brown rejected. In Brown, the Supreme Court stated that the legislation was an imposition of the government’s judgment, subject to a “parental veto” and “doubt[ed] that punishing third parties for conveying protected speech to children just in case their parents disapprove of that speech [was] a proper governmental means of aiding parental authority.” 564 U.S. at 802 & n.3. The interest Yost articulates in this case takes a subtly different angle. Ohio does not necessarily make assumptions about parents’ values or aim to enlist parents to prevent their Children from using social media. Rather Ohio seeks to involve parents because their involvement itself will inherently mitigate some of the harms of unsupervised social media use.

Finally, NetChoice argues that the Act does not only regulate contracting. But the purpose of regulating contracting is not invalid or unpersuasive just because the Act burdens speech as well. NetChoice does not explain why regulating contracting with Children would be a less than compelling state interest.

As for narrow tailoring, the Act need not be “perfectly drawn[,]” but Yost must show that the Act advances the government’s compelling interests “without being either too over-inclusive or under-inclusive.” Platt, 894 F.3d at 256 (quoting Winter v. Wolnitzek, 834 F.3d 681, 692 (6th Cir. 2016)). The former indicates “infringe[ment] on protected speech beyond the degree necessary” and “the latter may ‘reveal that a law does not actually advance a compelling interest No. 25-3371 NetChoice, LLC v. Yost Page 24

. . . .’” Id. (first citing Republican Party of Minnesota v. White, 536 U.S. 765, 775 (2002); and then quoting Williams-Yulee, 575 U.S. at 449).

Generally, NetChoice states that the Act is not sufficiently tailored because Ohio could instead give parents more information about the “many existing private tools to help them oversee their minor children online” and because the Act regulates an overbroad range of websites and protected speech. NetChoice Br. 57. Yost explains why informing parents about existing private tools for monitoring their children’s activity would not effectively prevent the harms that minors face. Children have at their disposal many ways of accessing the internet and can often do so anonymously, evading their parents’ supervision. Parents may not be able to keep up with what new social media platforms their children are using, in which case post hoc knowledge of available tools on a given platform is unhelpful. Prospectively requiring parental consent before a Child accesses a social media platform mitigates that problem. As for the Act’s range of covered sites and speech, Yost explains that the provisions that NetChoice complains are content-based actually narrow the scope of the Act to those sites that are likely to harm minors under the age of 16.

With respect to the government interest in child welfare, NetChoice argues that the Act is overbroad because it covers sites that do not employ the features that Yost says are harmful to children and underinclusive because it permits children to access the purportedly harmful sites if they have their parents’ consent. That the Act may reach some sites that do not use the features that Ohio finds harmful does not make it facially invalid, and NetChoice has not clarified which Members fall into that category. In fact, one has to wonder whether a site that does not use those features, such as push notifications or data-driven algorithms for displaying content, would even have constitutionally protected speech in play. Further, the Act should not be deemed fatally underinclusive simply because it permits Children to access potentially harmful platforms if they have parental consent. Again, the idea behind the Act seems to be that parental awareness and involvement will inherently mitigate the dangers of the platforms; Children will not be left to their own devices, so to speak.

On parental oversight, NetChoice contends that the Act is overinclusive because of the difficulty of verifying a parent-child relationship to process consent, because Yost has not shown No. 25-3371 NetChoice, LLC v. Yost Page 25

a real need among parents who wish to monitor their children’s online activity, and because nonconsent may not always reflect disapproval. First, verifying parental consent will not necessarily present such a great challenge. Second, parents do not know what they do not know. Part of the problem that the Act seeks to solve is exactly that lack of awareness among parents. That Yost has not shown an existing outcry from parents seeking greater control of their children’s internet use does not undermine the Act’s motivation to better involve parents. The Act strikes a balance between enlisting parents as a line of defense between Children and social media and overstepping into the realm of parent-child relationships. Parents who do not wish to monitor their Children are not obligated to do so beyond giving or withholding consent. And the Act is not overinclusive because parents may have irrelevant reasons for withholding consent; the Act would still promote involvement among those parents, which is a positive end unto itself.

On contracting, NetChoice says the Act is overinclusive because it requires the Operator to bar access entirely absent parental consent, rather than voiding or barring contracts. It is not clear that the Act actually requires the complete bar of access for a Child to an Operator’s site in the absence of the Operator’s requiring a contract. Yost stresses, and the language of the Act plausibly reflects, that the threshold criterion for an Operator to be subject to the Act’s requirements is that the Operator contracts with Children. “It has long been a tenet of First Amendment law that in determining a facial challenge to a statute, if it be ‘readily susceptible’ to a narrowing construction that would make it constitutional, it will be upheld.” Am. Booksellers, 484 U.S. at 397 (first citing Erznoznik v. City of Jacksonville, 422 U.S. 205 (1975); and then citing Broadrick, 413 U.S. 601). I would adopt Yost’s interpretation of the statute, which is that Operators may avoid penalties by doing away with their contracting requirements.

NetChoice additionally asserts that the Act is underinclusive because it does not cover all sites that require a contract for access. The Act targets the social media sites because of the aggregate threat they pose to minors. That the Act is not focused on a site like The New York Times does not call the state’s interest into question.

NetChoice attempts to undermine the Act by separating the multiple interests inherent in the Act’s objectives. But the convergence of those interests is what explains the Act’s design. In Ohio’s view, the benefits of the Covered Operators’ predatory contracting spur Covered No. 25-3371 NetChoice, LLC v. Yost Page 26

Operators to implement features that are dangerous for minors and avert parental oversight. The more that minors use the platforms, the more lucrative data Covered Operators can harvest and advertising they can sell. The Act works by promoting parental oversight as a hurdle to contracting. Ohio need not tailor the Act to each interest separately, in disregard of context.

At bottom, the Act imposes a parental consent requirement. That requirement constitutes a marginal burden that precisely targets the multi-faceted problem that Ohio has identified: Children’s unsupervised assent to terms and conditions for use of platforms that take advantage of and harm them. Parental consent will not always be narrowly tailored to the compelling interest in protecting minors’ wellbeing. It works here because the nature of the harm itself is that Children’s unsupervised use of social media puts them at risk of the adverse effects of prolonged and unregulated exposure. That premise differs from a case where parental consent is a tool to impose a value judgment about subject matter, subject to a parental veto. This opinion does not endorse states’ ability, for example, to censor information about American history or medical services. In this case, though, NetChoice has not shown that the Act’s parental-consent requirement violates its Members’ First Amendment rights, let alone that the Act violates the First Amendment on its face.

Judge Batchelder’s view, as I understand it, is that we should not reach the substantive merits of NetChoice’s First Amendment claim because NetChoice had to set up Moody’s facial challenge framework as a threshold matter. But the analysis is not necessarily required to proceed in that order. Because I do not think that NetChoice’s arguments prevail in establishing any unconstitutional applications of the Act, a fortiori NetChoice has not shown that the unconstitutional applications outweigh the constitutional ones, no matter the scope. In fact, in Moody, after establishing the path that the lower courts must follow, the Court proceeded to analyze the merits of the Texas statute in question and concluded that it would not pass First Amendment scrutiny. 603 U.S. at 726–40. Moody did not conduct that same analysis for the Florida statute, seemingly because the Eleventh Circuit had independently arrived at the conclusion that the Florida statute violated the First Amendment. See id. at 722. The Court remanded because, given that the laws had at least some unconstitutional applications, a full No. 25-3371 NetChoice, LLC v. Yost Page 27

facial analysis was appropriate. In this case, because NetChoice’s First Amendment arguments fail anyway, a fuller record of the Act’s scope would serve no purpose.

C. NetChoice’s Vagueness Challenge

The district court held that the Act’s coverage provisions are unconstitutionally vague. NetChoice, 778 F. Supp. 3d at 957–58. Specifically, it took issue with Section (B), which imposes the Act’s requirements on “operators that ‘target[ ] children’ or are ‘reasonably anticipated to be accessed by children[,]’” id. at 957 (first alteration in original) (quoting Ohio Rev. Code § 1349.09(B)), Section (C)’s “eleven-factor list that the Attorney General or a court may use to determine if a website is indeed covered” by Section (B), id. (quoting Ohio Rev. Code § 1349.09(C)), and Section (O)’s “exception for ‘established’ and ‘widely recognized’ media outlets whose ‘primary purpose’ is to ‘report news and current events,’” id. (quoting Ohio Rev. Code § 1349.09(O)(2)). Yost argues that the Act is not unconstitutionally vague because it uses plain, objective terms and provides a grace period for Covered Operators to correct mistakes. NetChoice maintains that the district court correctly granted summary judgment in its favor on this issue.

We review de novo the district court’s grant of summary judgment to NetChoice on its vagueness claim, “drawing all permissible inferences in favor of” Yost. Platt, 894 F.3d at 246 (citing Rose v. State Farm Fire & Cas. Co., 766 F.3d 532, 535 (6th Cir. 2014)). “[S]ummary judgment is appropriate when ‘there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.’” Byrd v. Tenn. Wine & Spirits Retailers Ass’n, 883 F.3d 608, 613 (6th Cir. 2018) (quoting Fed. R. Civ. P. 56(a)), aff’d sub nom. Tenn. Wine & Spirits Retailers Ass’n v. Thomas, 588 U.S. 504 (2019). No. 25-3371 NetChoice, LLC v. Yost Page 28

The “fundamental principle” undergirding constitutional vagueness challenges is that “laws which regulate persons or entities must give fair notice of conduct that is forbidden or required.” F.C.C. v. Fox Television Stations, Inc., 567 U.S. 239, 253 (2012). That requirement is integral to due process. Id.

[T]he void for vagueness doctrine addresses at least two connected but discrete due process concerns: first, that regulated parties should know what is required of them so they may act accordingly; second, precision and guidance are necessary so that those enforcing the law do not act in an arbitrary or discriminatory way.

Id. “Rigorous adherence” to those demands is essential when “ambiguity” threatens to “chill protected speech.” Id. at 253–54.

To prevail on its vagueness claim, NetChoice “must show either that the [Act] (1) ‘fail[s] to provide people of ordinary intelligence a reasonable opportunity to understand what conduct it prohibits’ or (2) ‘authorize[s] or even encourage[s] arbitrary and discriminatory enforcement.’” Platt, 894 F.3d at 246 (quoting Hill v. Colo., 530 U.S. 703, 732 (2000)). “While ‘a more stringent vagueness test should apply’ to laws abridging the freedom of speech, that standard is relaxed” where the challenged law “imposes civil rather than criminal penalties and includes an implicit scienter requirement.” Id. (quoting Vill. of Hoffman Ests. v. Flipside, Hoffman Ests., Inc., 455 U.S. 489, 499 (1982)). “[A] statute will be struck down as facially vague only if the plaintiff has demonstrate[d] that the law is impermissibly vague in all of its applications.” Libertarian Party of Ohio v. Husted, 751 F.3d 403, 422 (6th Cir. 2014) (alterations in original) (quoting Green Party of Tenn. v. Hargett, 700 F.3d 816, 825 (6th Cir. 2012)).

We look to the language of the Act to assess whether it provides sufficient notice of what it proscribes. Platt, 894 F.3d at 246 (quoting Grayned v. City of Rockford, 408 U.S. 104, 110 (1972)). “[M]athematical certainty” is not required, and “flexibility and reasonable breadth” are not fatal to a statute’s clarity. Id. at 246–47 (quoting Grayned, 408 U.S. at 110). If the words in the statute are commonly understood, they need not be defined. Id. at 247 (quoting United States v. Hollern, 366 F. App’x 609, 612 (6th Cir. 2010)). If, however, the Act “forbids or requires the doing of an act in terms so vague that men of common intelligence must necessarily guess at its No. 25-3371 NetChoice, LLC v. Yost Page 29

meaning and differ as to its application, [it] violates . . . due process of law.” McGlone v. Cheek, 534 F. App’x 293, 297 (6th Cir. 2013) (quoting Connally v. Gen. Const. Co., 269 U.S. 385, 391 (1926)). The U.S. Supreme Court has struck down statutes that contain “wholly subjective judgments without statutory definitions, narrowing context, or settled legal meanings.” United States v. Williams, 553 U.S. 285, 306 (2008).

The stricter First Amendment standard applies to the Act, given that it implicates at least some Members’ protected speech and that the burden of compliance or fear of repercussions could chill that speech. See McGlone, 534 F. App’x at 298 (“[U]ncertain meanings inevitably lead citizens to ‘steer far wider of the unlawful zone’ . . . than if the boundaries of the forbidden areas were clearly marked.” (quoting Grayned, 408 U.S. at 109)). However, given that the Act provides for only civil penalties and not criminal punishment, the standard is somewhat relaxed.

For some Operators, a person of ordinary intelligence might not understand what the Act requires, and determining whether they fall within the coverage definition will be a subjective and unpredictable endeavor. There is no bright line between the subject matter, advertisements, or celebrities, for example, that draw the attention of Children and those that do not. As NetChoice argues, the Act cover sites that are reasonably anticipated to be accessed by Children, based on optional “vague, subjective, or [] broad” factors. NetChoice Br. 72. Similarly, the exemption of “established and widely recognized media outlet[s]” might be vague as applied to some hypothetical Operators. Ohio Rev. Code § 1349.09(O).

Yost points out that the Act’s factors track those that appear in the regulations under the federal Children’s Online Privacy Protection Act (COPPA), 15 U.S.C. §§ 6501–06, which requires operators of sites that are directed to children under 13 years of age and that collect personal information from children to obtain verifiable parental consent before doing so. 15 U.S.C. §§ 6501(1), 6502(b)(1)(A). A regulation under COPPA instructs the Federal Trade Commission to determine whether a site is directed to children based on criteria that match the Act’s factors almost exactly. The list includes:

subject matter, visual content, use of animated characters or child-oriented activities and incentives, music or other audio content, age of models, presence of child celebrities or celebrities who appeal to children, language or other No. 25-3371 NetChoice, LLC v. Yost Page 30

characteristics of the website or online service, . . . advertising[, and] . . . empirical evidence . . . .

16 C.F.R. § 312.2. But Yost does not explain why the resemblance of the Act to federal law makes the Act less vague. Furthermore, the list of factors in the Act is non-exhaustive, so the Attorney General and the courts might even consider other factors an Operator could not predict. Therefore, in some instances, the Act’s language might go beyond flexibility and commonly understood terms to the point where further definition would be necessary to understand whether the Act applies. In those instances, the Act may be vague. For many Operators, though, the coverage provisions will not be vague at all. In fact, it is unlikely that an Operator within the scope of the Act’s threshold definition would find Section (B), (C), or (O) to be vague in application. According to Section (A)(1), the Act reaches only those Operators of products that facilitate social interaction and creation of personal profiles, networks, and shareable content. In a phrase, it covers social media companies. NetChoice offers no example of an Operator that is covered under Section (A)(1) and yet credibly uncertain as to whether it is reasonably anticipated to be accessed by Children, as provided in Sections (B) and Section (C).

Several Members that believe the Act regulates them deliberately target Children, mitigating any alleged vagueness in Sections (B) and (C). For example, Goodreads, a “site for readers and book recommendations[,]” candidly anticipates use by Children. Roin Decl. Supp. Pl.’s Mot. TRO & Prelim. Inj., R. 2-3, PageID #88. It “facilitates discussion . . . among teen readers (including teens between 13 and 16)” and features the young adult genre of books “written for, published for, or marketed to adolescents and young adults, roughly ages 13 to 18.” Id. at PageID #89 (citation omitted). Operators like TikTok and YouTube, which have specific services and policies for minors younger than 13, clearly anticipate being accessed by Children. None of those Operators can seriously maintain that they are unsure whether they are reasonably anticipated to be accessed by Children.

Similarly, an Operator whose purpose is to report news and current events may struggle to understand whether it is sufficiently “established and widely recognized” to be exempt under Section (O), but any such borderline site that comes to mind would not pass the threshold No. 25-3371 NetChoice, LLC v. Yost Page 31

coverage description of Section (A)(1). To use specific examples from the record, Breitbart or the Onion might not know whether they are excluded under Section (O), but they do not seem to be covered as social media Operators under Section (A)(1) in any event. Conversely, X (formerly Twitter) and Facebook might consider themselves to be news sources, but they would not be exempt under Section (O) because “interaction between users is [not] limited to . . . [incidental c]omments . . . .” Id. § 1349.09(O). And the exclusion certainly would not be vague for an Operator whose purpose is unrelated to news and current events, like Dreamwidth.

NetChoice’s other point about vagueness is that the Act delegates too much discretion to the Attorney General. To evaluate that argument, we consider the Act “collectively” to determine whether it “provide[s] explicit standards guiding [its] enforcement.” Platt, 894 F.3d at 252 (alterations in original) (quoting United Food & Com. Workers Union, Local 1099 v. Sw. Ohio Reg’l Transit Auth., 163 F.3d 341, 359 (6th Cir. 1998)). If the Act “fails to constrain” the Attorney General’s “‘decision to limit speech’ with ‘objective criteria[, it]’ is unconstitutionally vague.” Miller v. City of Cincinnati, 622 F.3d 524, 539 (6th Cir. 2010) (quoting United Food & Com. Workers, 163 F.3d at 359). For example, a prohibition on advertisements that are “controversial” or not “aesthetically pleasing” may be unconstitutionally vague “without any further guidance” because it gives officials discretion to disallow advertisements based on their subjective opinions. Id. at 539–40 (quoting United Food & Com. Workers, 163 F.3d at 359–60).

The Act’s language and enforcement mechanisms risk arbitrary or discriminatory enforcement to some extent. The Attorney General has authority to investigate an Operator if he “has reason to believe” it is noncompliant. Ohio Rev. Code § 1349.191(B). He then may bring a civil action “[i]f it appears that an operator . . . failed to comply . . . .” Id. § 1349.09(H). Those enforcement provisions give the Attorney General significant discretion. But that discretion is curtailed because a court will only impose civil penalties for noncompliance if the “court finds that” a Covered Operator “entered into a contract with a child without consent of the child’s parent or guardian . . . .” Id. § 1349.09(I). Even the costs of cooperating with the Attorney General’s investigation, though, such as the time and expense of producing witnesses and evidence, all at the discretion of the Attorney General, raise some concern. No. 25-3371 NetChoice, LLC v. Yost Page 32

Yost argues that where an Operator is confused about the requirements of the Act, the safe harbor provision mitigates any risk of mistake. But the safe harbor is not all that helpful. The provision states, “If an operator is in substantial compliance with this section, the attorney general shall provide written notice to the operator before commencing a civil action” and give the Operator 90 days to “[c]ure[] the violation” and “[p]rovide[] the attorney general with written documentation that the violation has been cured and that the operator has taken measures sufficient to prevent future violations.” Id. § 1349.09(M). But that provision only saves an Operator that is substantially compliant, which would mean that the Operator understood that it was covered and undertook the efforts mandated by the Act. Furthermore, “substantial compliance” is undefined. A person of ordinary intelligence would not necessarily understand how compliant an Operator would need to be to avoid being haled into court.

Again, in some cases, discretion is not an issue. The Act’s standards are not entirely open-ended, free-floating value judgments. They do not ask whether the Operator’s content is “controversial” or “aesthetically pleasing.” Although perhaps blurry at the edges, the objective standard of whether the Operator targets or is reasonably likely to be accessed by Children tethers the Attorney General’s inquiry. For a Covered Operator that clearly targets Children and clearly has not implemented the Act’s requirements, the Act is not unconstitutionally vague. That Operator would clearly fit within the criteria regardless of the Attorney General’s subjective understanding of the full breadth of the Act’s coverage. NetChoice has not shown that the Act is impermissibly vague in all of its applications or even in its application to all of the Members. Therefore, summary judgment in favor of NetChoice on its void-for-vagueness claim was improper. See Platt, 894 F.3d at 251 (“[A]lmost any criminal or civil prohibition is susceptible to clever hypotheticals testing its reach.”).

III. CONCLUSION

NetChoice does not have third-party standing to assert the rights of its Members’ minor users and has otherwise failed to show why Ohio’s Parental Notification by Social Media Operators Act, Ohio Rev. Code § 1349.09, facially violates the U.S. Constitution. The Act is content-based and deserves strict scrutiny but is narrowly tailored to compelling state interests and is not impermissibly vague. Judge Batchelder agrees that NetChoice lacks third-party No. 25-3371 NetChoice, LLC v. Yost Page 33

standing and that NetChoice has failed to meet its burden on the constitutional challenges, but would resolve the facial First Amendment challenge based on Moody, without reaching the merits. Judge Ritz thinks that NetChoice does have third-party standing and that the Act fails strict scrutiny, and therefore he does not reach the vagueness issue. Given Judge Batchelder’s and my limited consensus that Yost has prevailed, we REVERSE the district court’s judgment and REMAND with instructions to enter judgment in favor of Defendant Yost. We therefore need not reach the issue of the scope of the district court’s injunction. No. 25-3371 NetChoice, LLC v. Yost Page 34

CONCURRENCE _________________

ALICE M. BATCHELDER, Circuit Judge, concurring in the judgment. In 2023, Ohio passed the Parental Notification by Social Media Operators Act (“The Act”), codified at Ohio Rev. Code § 1349.09, in response to what it considered to be ongoing, substantial harm to minors from their social-media use. At root, The Act requires that a covered website operator must obtain parental consent before any unemancipated minor under the age of sixteen may “register, sign up, or otherwise create a unique username to access or utilize” the website. See id. § 1349.09(B)(1). NetChoice, LLC, a trade association including several prominent social-media operators, sued Ohio Attorney General David Yost to enjoin his enforcement of The Act. NetChoice contends that The Act facially abridges the freedom of speech and facially violates the “void for vagueness” doctrine.

The district court granted summary judgment in NetChoice’s favor and permanently enjoined enforcement of The Act. NetChoice, LLC v. Yost, 778 F. Supp. 3d 923, 959 (S.D. Ohio 2025). Finding that The Act restricted speech and applying strict scrutiny, the court held that The Act violated the First Amendment rights of NetChoice’s members and NetChoice’s members’ potential minor users. Id. at 947–57. The district court also held that The Act’s purported vagueness ran afoul of the Fourteenth Amendment’s Due Process Clause. Id. at 957– 58.

The record below supports neither conclusion. First, NetChoice failed to satisfy its burden in raising a facial challenge to The Act. See Moody v. NetChoice, LLC, 603 U.S. 707, 725 (2024). Second, NetChoice’s facial-vagueness challenge fails because the purportedly vague provisions are either not vague or not vague as applied to NetChoice’s members.

My colleagues’ opinions apply First Amendment strict scrutiny to The Act, though they disagree on whether the Act survives that level of scrutiny. While I seriously question whether strict scrutiny is appropriate in this case, I do not reach the merits of NetChoice’s First Amendment claim because NetChoice does not meet its prerequisite burden in facially No. 25-3371 NetChoice, LLC v. Yost Page 35

challenging The Act. And while I agree with Judge Clay that NetChoice’s facial-vagueness challenge fails (Judge Ritz does not reach the issue), I do so with different reasoning. So, I join in the judgment to REVERSE the district court’s judgment and REMAND with instruction to enter judgment in favor of Yost.

I.

In the proceedings below, the parties “agreed to forego discovery,” agreed “not [to] rely on expert testimony,” and “stipulate[d] that the evidence . . . submitted with their respective [preliminary-injunction] briefing . . . is admissible.” Yost, 778 F. Supp. 3d at 959. I therefore consider the stipulated factual record from the preliminary-injunction stage, as well as the parties’ supplemental materials from their summary-judgment briefing.

A. Social Media and Minors

When Ohio passed The Act in 2023, it joined a growing list of states concerned about their minor residents’ wellbeing as social media becomes an ever more pervasive part of those minors’ lives.1 Yost identifies three main categories of harm to minors connected to social- media use: (1) poor health and behavioral outcomes, including increased rates of depression, suicide, self-harm, bullying, eating disorders, academic decline, and addiction; (2) online and offline sexual abuse; and (3) onerous, privacy-minimizing contractual relationships between minors and social-media companies. See Aff. of Ohio Lt. Gov. Jon Husted, R. 42-5, PageID 534. Yost asserts a “compelling interest in protecting minors from these and all other harmful effects” of social-media use, as well as a compelling interest “in protecting parents’ rights and abilities to make decisions that are in the best interest of their children.” Id.

1 See, e.g., NetChoice, LLC v. Fitch, 606 U.S. ---, 145 S. Ct. 2658 (2025) (Mississippi); Computer & Commc’ns Indus. Ass’n v. Uthmeier, No. 25-11881, 2025 WL 3458571 (11th Cir. Nov. 25, 2025) (Florida); NetChoice, LLC v. Bonta (Bonta II), 170 F.4th 744, (9th Cir. 2026) (California); Computer & Commc’ns Indus. Ass’n v. Paxton, 747 F. Supp. 3d 1011 (W.D. Tex. 2024) (Texas); NetChoice, LLC v. Reyes, 748 F. Supp. 3d 1105 (D. Utah 2024) (Utah); NetChoice, LLC v. Griffin, No. 5:23-CV-5105, 2025 WL 978607 (W.D. Ark. Mar. 31, 2025) (Arkansas); NetChoice v. Skrmetti, No. 3:24-CV-01191, 2025 WL 1710228 (M.D. Tenn. June 18, 2025) (Tennessee); NetChoice v. Carr, 789 F. Supp. 3d 1200 (N.D. Ga. 2025) (Georgia); NetChoice v. Weiser, 808 F. Supp. 3d 1223 (D. Colo. 2025) (Colorado); NetChoice v. Brown, No. CV RDB-25-0322, 2025 WL 3267786 (D. Md. Nov. 24, 2025) (Maryland); NetChoice v. Murrill, No. CV 25-231-JWD-RLB, 2025 WL 3634112 (M.D. La. Dec. 15, 2025) (Louisiana); NetChoice, v. Jones, No. 1:25-CV-2067 (PTG/LRV), 2026 WL 561099 (E.D. Va. Feb. 27, 2026) (Virginia). No. 25-3371 NetChoice, LLC v. Yost Page 36

By way of proof, Yost points in part to a 2023 U.S. Surgeon General’s Advisory entitled “Social Media and Youth Mental Health” (the “SG Advisory”). R. 42-2, PageID 490. Describing “the effects of social media on youth mental health” as a “significant public health challenge[] that require[s] the nation’s immediate awareness and action,” the SG Advisory relies heavily on “meta-analyses and systemic literature reviews.” Id., PageID 492. Some of the SG Advisory’s warnings are pure statistical assessments: “Up to 95% of youth ages 13–17 report using a social media platform, with more than a third saying they use social media ‘almost constantly.’” Id., PageID 493. “[N]early 40% of children ages 8–12 use social media.” Id. “[A]ccording to a survey of 8th and 10th graders, the average time spent on social media is 3.5 hours per day, 1-in-4 spend 5+ hours per day[,] and 1-in-7 spend 7+ hours per day on social media.” Id., PageID 499. “On a typical weekday, nearly 1-in-3 adolescents report using screen media until midnight or later.” Id. Prevalent social-media use by minors is also reflected in other research articles proffered by Yost. One study estimates that, in 2022, YouTube had 31.5 million users under the age of 13 and 18.4 million users aged 13 to 17. Raffoul Study, R. 42-1, PageID 484. In the same year, Twitter and Facebook had an estimated 7 million and 9.9 million users under the age of 18, respectively. Id. The same study estimates that “Instagram, Snapchat, and TikTok had an average of nearly 18 million users under the age of 18 years.” Id.

Beyond these statistics, Yost presents evidence linking social-media use and adverse mental and behavioral health outcomes. The SG Advisory identifies multiple studies to this effect: See, e.g., R. 42-2, PageID 495 (“A longitudinal cohort study of U.S. adolescents aged 12– 15 (n=6,595) that adjusted for baseline mental health status found that adolescents who spent more than 3 hours per day on social media faced double the risk of experiencing poor mental health outcomes including symptoms of depression and anxiety.”). The kinds of research analyzed by the Surgeon General range from correlation studies, see id., PageID 496 (citing “a study conducted among 14-year-olds (n = 10,904) [that] found that greater social media use predicted poor sleep, online harassment, poor body image, low self-esteem, and higher depressive symptom scores with a larger association for girls than boys”), to surveys, see id., PageID 498–99 (discussing youth surveys indicating that “one-third or more” of “girls aged 11– 15” “say they feel ‘addicted’ to a social media platform” and that “[o]ver half of teenagers report that it would be hard to give up social media”), to randomized, controlled experiments, see id., No. 25-3371 NetChoice, LLC v. Yost Page 37

PageID 496 (citing a “randomized controlled trial among young adults and adults [that] found that deactivation of a social media platform for four weeks improved subjective well-being . . . by about 25–40% of the effect of psychological interventions like self-help therapy, group training, and individual therapy”). Yost’s record provides evidence that “frequent and problematic social media use” is linked to “changes in brain structure similar to changes seen in individuals with substance use or gambling addictions.” Id., PageID 498. Perhaps most strikingly, the unique staggered introduction of Facebook across domestic colleges allowed researchers to analyze a vast amount of natural-experiment data relating to social-media use and mental health: “the roll-out of the platform was associated with an increase in depression (9% over baseline) and anxiety (12% over baseline) among college-aged youth (n = 359,827 observations).” Id., PageID 496.

This is not to say the record before us universally points to a causal relationship between children’s using social media and then manifesting symptoms of poor mental health. As the SG Advisory concedes, “[m]ost prior research to date has been correlational, focused on young adults or adults, and generated a range of results.” Id., PageID 500. Another study proffered by Yost paints a similar picture:

Yet there is still considerable uncertainty about how social media use relates to well-being. Meta-analyses have identified small or negligible negative links between social media use and well-being, while experimental evidence is mixed. Longitudinal observational studies that have investigated the predictive relationships between social media use and well-being have found that they are either reciprocal, only present in a certain direction or sex or not present at all. The lack of concrete evidence is an issue routinely highlighted by academics, medical professionals, and policymakers.

R. 42-3, PageID 517. As this study goes on to posit, some of the muddied data could be the result of differing susceptibilities of children based on age and sex. See id., PageID 518 (stratifying “life satisfaction ratings” based on hours of daily social-media use by age and sex). This study, as well as the SG Advisory, also proposes that the data may be obscured by a possible “Goldilocks” relationship, whereby limited, supervised social-media use may be more beneficial to some older youth than would be none at all, but that usage beyond some threshold then produces deleterious effects. See id., PageID 517–518; see also SG Advisory, R. 42-2, No. 25-3371 NetChoice, LLC v. Yost Page 38

PageID 500 (noting a “potentially bidirectional” relationship between social-media use and mental health in minors).

NetChoice challenges the weight of Yost’s evidence, arguing that “‘nearly all of the research’ showing any harmful effects here ‘is based on correlation.’” Appellee NetChoice’s Br. at 54 (quoting Brown v. Ent. Merchants Ass’n, 564 U.S. 786, 800 (2011)). The district court agreed with that characterization of Yost’s record. Yost, 778 F. Supp. 3d at 955–56. Further, NetChoice emphasizes that the SG Advisory itself concedes that “[s]ocial media can provide benefits for some youth by providing positive community and connection with others who share identities, abilities, and interests.” Appellee NetChoice’s Br. at 55 (emphasis omitted) (quoting R.42-2, PageID 495).

But even accepting that this evidence of correlation does not amount to proof of causation, two factual issues undermine NetChoice’s point. First, a major hurdle to studying the relationship between social-media use and minors’ mental-health outcomes resides with the social-media companies themselves: The SG Advisory maintains that “[t]here is broad concern among the scientific community that a lack of access to data and lack of transparency from technology companies have been barriers to understanding the full scope and scale of the impact of social media on mental health and well-being.” R. 42-2, PageID 500. Second, one particular adverse effect of minors’ use of social media—addiction—is an intentional element of some social-media platforms’ designs. See Appellant Yost’s Br. at 5–6; see also Yost’s Mot. Summ. J., R. 42, PageID 441–42. Yost points us to the Congressional testimony of a former executive of one of NetChoice’s largest members: “[W]e sought to mine as much human attention as possible and turn it into historically unprecedented profits. To do this, we didn’t simply create something useful and fun. We took a page from Big Tobacco’s playbook, working to make our offering addictive at the outset.” Id., PageID 441 (citation omitted). Yost points out that his excessive-use-by-design inference is borne out by the financial incentives undergirding social- media product design. See id., PageID 441 (“The obvious answer to this [continued-growth] problem is that these companies must capture a larger share of their billions of users’ time and attention. This incentivizes them, now more than ever, to make their services addictive. Their continued growth depends on it. And it’s working.”). One study estimates that the annual No. 25-3371 NetChoice, LLC v. Yost Page 39

advertising revenue in 2022 for Facebook, Instagram, Snapchat, TikTok, Twitter (now X), and YouTube from children under 13 was over $2 billion. Raffoul Study, R. 42-1, PageID 484–485. For minors under 18, that value is estimated at nearly $11 billion. Id. Notably, NetChoice does not counter these allegations that social media is designed to be addictive to youth.

Yost contends that minors face other risks from social-media use, including unsupervised online interactions leading to encounters with sex predators. Yost proffers an affidavit from an investigator of online child-exploitation crimes who testified that social media’s prevalent features (anonymity, easy ability to interact with other users, easy ability to review other users’ previous posts, end-to-end encrypted private messaging, lack of mandatory parental supervision) make the platforms ideal vectors for grooming and exploiting minors. R. 42-4, PageID 528–30.

Yost also identifies many of social-media operators’ contractual terms of service as harmful to minors, both as an independent harm and as an augmentation of other, previously discussed harms. In particular, Yost alleges that these terms of service impose agreement-to- liability, arbitration, choice-of-law, forum-selection, and intellectual-property provisions that favor the social-media providers at the expense of minor users. See, e.g., Yost’s Mot. Summ. J., R. 42, PageID 450–53 (analyzing various platforms’ terms of service). According to Yost, these terms hamper the minors’ privacy and intellectual-property interests and make it more difficult for them to pursue relief related to other harms, such as adverse mental-health outcomes.

NetChoice does not refute the alleged harms other than to assert that Yost fails to establish causation connecting minors’ social-media use and adverse health and behavioral outcomes. NetChoice instead points us to the potential benefits of social-media use and existing parental supervision tools. According to NetChoice, and as admitted by Yost’s proffered SG Advisory, social media can provide minors with communities of like-minded and like-interested users. Appellant NetChoice’s Br. at 55 (citation omitted). And parents can control which electronic devices their children access, whether those devices are internet-enabled, and when children may access them. NetChoice alleges that many devices, networks, and software systems come with built-in parental-control options, allowing parents to limit apps, features, content, and usage, and that many NetChoice members have developed their own suite of protections and controls, including content moderation. Id. at 5–7. No. 25-3371 NetChoice, LLC v. Yost Page 40

In sum, as the district court recognized, “good things [may] be said about social media platforms,” but minors’ social-media use “come[s] with a cost” to minors and a multi-billion- dollar boon to social-media operators. See Yost, 778 F. Supp. 3d at 932–33. Social-media operators “do not appear to be incorporating user-agnostic judgments when promoting content” to adults or minors. Id. at 933. Rather, “[s]ocial media platforms are highly incentivized to keep youth online—children and adolescents’ online experiences are heavily monetized through advertising revenue on platforms’ websites and mobile applications.” Raffoul Study, R. 42-1, PageID 483. Platforms indisputably employ features and algorithms “that leverage user data to serve content recommendations” and “maximize engagement.” SG Advisory, R. 42-2, PageID 498. And while Yost’s evidence of harm may be primarily “based on correlation,” see Yost, 778 F. Supp. 3d at 955, that correlation is significant enough to garner the attention of researchers, the U.S. Surgeon General, and numerous state legislatures and governors, among others. NetChoice does not refute this harm. Rather, NetChoice contends that the purported harm is insufficient justification for The Act as a matter of constitutional scrutiny.

B. The Parental Notification by Social Media Operators Act

In response to this alleged harm, Ohio’s elected representatives passed and its Governor signed The Act. The legislation applies to “operator[s] of an online website, service, or product that targets children, or is reasonably anticipated to be accessed by children.” Ohio Rev. Code § 1349.09(B). An “operator” is defined as “any business, entity, or person that operates an online web site, service, or product that has users in [Ohio] and that allows those users to . . . (a) [i]nteract socially with other users within the confines of the online web site, service, or product; (b) [c]onstruct a public or semipublic profile for the purpose of signing into and using the online web site, service, or product; (c) [p]opulate a list of other users with whom an individual shares or has the ability to share a social connection within the online web site, service, or product; [and] (d) [c]reate or post content viewable by others, including on message boards, chat rooms, video channels, direct or private messages or chats, and a landing page or main feed that presents the user with content generated by other users.” Id. § 1349.09(A)(1).

Excluded from The Act’s coverage are websites, services, and products in which the “predominant or exclusive function” is “cloud storage or cloud computing services,” “broadband No. 25-3371 NetChoice, LLC v. Yost Page 41

internet access services,” or “search engine services,” except “with respect to content and communications created or controlled by” the website’s, service’s, or product’s provider, affiliate, or subsidiary. Id. § 1349.09(N). Also excluded are websites, services, or products respecting which “interaction[s] between users” consist only of “[r]eviewing products offered for sale by electronic commerce or commenting on reviews posted by other users” or “[c]omments incidental to content posted by an established and widely recognized media outlet, the primary purpose of which is to report news and current events.” Id. § 1349.09(O).

The Act provides a list of factors that “the attorney general or a court may consider” when determining whether a website, service, or product targets, or is reasonably anticipated to be accessed by, children: (1) Subject matter; (2) Language; (3) Design elements; (4) Visual content; (5) Use of animated characters or child-oriented activities and incentives; (6) Music or other audio content; (7) Age of models; (8) Presence of child celebrities or celebrities who appeal to children; (9) Advertisements; (10) Empirical evidence regarding audience composition; and (11) Evidence regarding the intended audience. Id. § 1349.09(C). Children are defined as unemancipated minors under the age of 16. Id. § 1349.09(A)(2). The statute is silent on age verification, but at oral argument on summary judgment, counsel for Yost took the position that “[t]his is a self-reporting statute.” R. 63, PageID 1399 (“[I]f the child fibs and says that they’re over the age of 16, the Act is not requiring these companies to go through and verify the age of all of its users.”).

The Act imposes three requirements on covered operators. First, they must “[o]btain verifiable consent for any contract made with a child, including terms of service, to register, sign up, or otherwise create a unique username to access or utilize the online web site, service, or product, from the child’s parent or legal guardian using” one of five approved methods. Id. § 1349.09(B)(1). Second, they must “[p]resent to the child’s parent or legal guardian a list of the features offered by an operator’s online web site, service, or product related to censoring or moderating content, including any features that can be disabled for a particular profile.” Id. § 1349.09(B)(2). Third, they must “[p]rovide to the child’s parent or guardian a web site link at which the parent or legal guardian may access and review [that] list of features.” Id. § 1349.09(B)(3). Parents must be permitted to withdraw consent. Id. § 1349.09(F). Any child No. 25-3371 NetChoice, LLC v. Yost Page 42

lacking affirmative consent from his or her parent or legal guardian must be denied access to the website, service, or product. Id. § 1349.09(E).

The Ohio Attorney General is charged with investigating violations and enforcing The Act. Id. § 1349.09(G)–(H). A covered operator’s failure to comply can result in significant civil monetary penalties and responsibility to pay the Attorney General’s investigating and litigating costs. See id. § 1349.09(I)–(K).

C. NetChoice’s Challenge

NetChoice is a trade association of internet companies. Its members operate several websites that NetChoice claims are covered by The Act: Google’s YouTube; Meta’s Facebook, Instagram, and Threads; Pinterest; X; Nextdoor; and Dreamwidth. Its members also include Snap, Inc. (operator of SnapChat). NetChoice represents that “some of [its] members own or operate what are commonly referred to as ‘social media’ websites.” Aff. of NetChoice Vice President and Gen. Couns. Carl Szabo, R. 2-1, PageID 64. According to NetChoice, “each covered member’s website publishes, disseminates, creates, curates, and distributes protected speech.” Id., PageID 65.

In 2024, NetChoice brought this suit to stop enforcement of The Act. Seeking declaratory and injunctive relief under 28 U.S.C. § 2201 and 42 U.S.C. § 1983, NetChoice argues that The Act abridges the freedom of speech enshrined in the First Amendment and applied against the states by the Fourteenth Amendment to the United States Constitution. The district court granted NetChoice’s motions for a temporary restraining order and preliminary injunction. After considering the parties’ cross-motions for summary judgment, the district court granted NetChoice’s motion, denied Yost’s motion, and enjoined all enforcement of The Act. Yost, 778 F. Supp. 3d at 959. Yost appeals from that judgment, contending that (1) NetChoice lacks standing to assert the rights of its members’ potential minor users, (2) NetChoice’s facial First Amendment challenge fails under any level of constitutional scrutiny, (3) NetChoice’s facial-vagueness challenge fails based on a plain reading of the text, and (4) the district court’s relief was overly broad. No. 25-3371 NetChoice, LLC v. Yost Page 43

II.

This court reviews de novo a grant of summary judgment, including for lack of standing. Kareem v. Cuyahoga Cnty. Bd. of Elections, 95 F.4th 1019, 1021 (6th Cir. 2024) (citation omitted). “Summary judgment is proper when the ‘movant shows that there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.’” Id. (citing Fed. R. Civ. P. 56(a)). When this court addresses cross-motions for summary judgment, it evaluates each party’s motion on its own merits, drawing all reasonable inferences against the party whose motion is under consideration. Lyngaas v. United Concordia Companies, Inc, 140 F.4th 749, 751 (6th Cir. 2025), reh’g denied sub nom. Brian J. Lyngaas, D.D.S., P.L.L.C. v. United Concordia Companies, Inc., No. 24-1777, 2025 WL 2214203 (6th Cir. July 8, 2025).

A. NetChoice’s Standing

1. Associational Standing

It is axiomatic that NetChoice must have standing to bring its claims. Our “threshold question” is whether NetChoice has “standing to sue under Article III of the Constitution.” See Food & Drug Admin. v. All. for Hippocratic Med., 602 U.S. 367, 378 (2024). NetChoice asserts “associational standing to raise its member websites’ rights” in satisfaction of our Article III standing requirements. Appellee NetChoice’s Br. at 15. Our recent caselaw has questioned whether associational standing remains congruent with the “irreducible constitutional minimum” of Article III standing given its lack of historical pedigree and the Supreme Court’s recent skepticism toward “prudential” standing doctrines intertwined with the associational standing test. Ass’n of Am. Physicians & Surgeons v. U.S. Food & Drug Admin., 13 F.4th 531, 537–542 (6th Cir. 2021). But without clearer guidance from the Court, this panel is nonetheless bound to follow “the current test for this type of proxy suit. An organization may sue on behalf of its members if it shows that: (1) its ‘members would otherwise have standing to sue in their own right’; (2) the ‘interests’ that the suit ‘seeks to protect are germane to the organization’s purpose’; and (3) ‘neither the claim asserted nor the relief requested requires the participation of individual members in the lawsuit.’” Id. at 537 (internal citation omitted) (quoting Hunt v. Wash. State Apple Advert. Comm’n, 432 U.S. 333, 343 (1977)). Yost does not challenge No. 25-3371 NetChoice, LLC v. Yost Page 44

NetChoice’s associational standing to raise its member’s rights, and nothing from the record suggests NetChoice fails to satisfy the test articulated in Hunt with regard to these two facial challenges on behalf of its members.

2. Third-Party Standing

NetChoice also asserts “prudential standing to raise the interests of members’ users.” Appellee NetChoice’s Br. at 15. In doing so, NetChoice essentially asks to “‘borrow’ its . . . members’ third-party standing to assert the separate rights of” its members’ users—or, more accurately, its potential minor users who stand to have their access to social media hampered by The Act. See Ass’n of Am. Physicians & Surgeons, 13 F.4th at 547. This raises two questions: Can NetChoice’s members assert third-party standing on behalf of their potential minor users? And can NetChoice then “stack” its members’ third-party standing upon its associational standing (a practice which one of our sister circuits labels “derivative standing,” see Pennsylvania Psychiatric Soc. v. Green Spring Health Servs., Inc., 280 F.3d 278, 291 (3d Cir. 2002))? Regarding the second question, at least one Supreme Court decision seems to countenance derivative standing. See New York State Club Ass’n, Inc. v. City of New York, 487 U.S. 1, 9 (1988) (permitting a “consortium of . . . private clubs and associations” to sue on behalf of its members based on those members’ “standing to bring this same suit on behalf of their own individual [private club] members”). But the answer to the first question obviates the need to definitively answer the second: NetChoice’s members cannot bring this lawsuit to vindicate the free-speech rights of minors. Minors would have to bring their own case should they seek to vindicate their rights.

To be sure, the framework of our third-party standing doctrine is hardly a model of clarity. The Supreme Court has alternatively articulated the doctrine as being “closely related to the question whether a person in the litigant’s position will have a right of action on the claim,” Lexmark Int’l, Inc. v. Static Control Components, Inc., 572 U.S. 118, 127 n.3 (2014) (quoting U.S. Dep’t of Lab. v. Triplett, 494 U.S. 715, 721, n. ** (1990)), suggesting an Article III link, or as being “an element of ‘prudential standing,’” id. (quoting Kowalski v. Tesmer, 543 U.S. 125, 128–129 (2004)). And at least one Justice has questioned third-party standing altogether. All. for Hippocratic Med., 602 U.S. at 397 (Thomas, J., concurring). No. 25-3371 NetChoice, LLC v. Yost Page 45

Regardless of the “doctrine’s proper place in the standing firmament,” Lexmark, 572 U.S. at 127 n.3., NetChoice’s gambit to assert minors’ rights falls short of our current third-party standing test: “To establish such third-party standing absent statutory authorization, plaintiffs must show both that they have a ‘“close” relationship’ with the person that possesses the right and that some ‘hindrance’ limits that person’s ability to assert it.” Digital Media Sols., LLC v. S. Univ. of Ohio, LLC, 59 F.4th 772, 782 (6th Cir. 2023) (quoting Kowalski, 543 U.S. at 130). NetChoice lacks a sufficiently close relationship with its potential minor members. In fact, NetChoice’s interests are often opposed to the interests of those it seeks to represent.

One of the primary harms alleged by Yost arises from the allegedly onerous terms of service imposed on minor users by social-media operators, including NetChoice’s members. Yost provides examples: Facebook (a platform operated by NetChoice member Meta) requires each of its users to allow Facebook to “transfer, store and distribute content and data to [its] data centers, partners, service providers, vendors and systems around the world, including outside [the user’s] country of residence.” Yost’s Mot. Summ. J., R. 42, PageID 450 (first alteration added) (citation omitted). Facebook’s users must also grant it a “non-exclusive, transferable, sub- licensable, royalty-free, and worldwide license to host, use, distribute, modify, run, copy, publicly perform or display, translate, and create derivative works of [the user]’s content.” Id. (alteration in original) (citation omitted). NetChoice member Dreamwidth’s terms allegedly contain an “indemnity/hold-harmless provision, liability limitations, warranty disclaimers, [a] choice-of-law/forum-selection provision, [and a] license to user-submitted content.” Id., PageID 452.

According to Yost, these terms of service directly benefit operators to the detriment of their minor users. In fact, The Act seeks to mitigate that harm through mandatory parental consent to those terms and, by extension, those parents’ now-informed supervision of their children’s social-media activities. Further, The Act seeks to protect those minor users from the potential adverse effects of social-media use. In sum, this is a consumer-protection law, and a consumer’s interests vis-à-vis such legislation are not the same as—and are often diametrically opposed to—a seller’s interests. That is the case here, and “[a] failure to satisfy [Kowalski’s] No. 25-3371 NetChoice, LLC v. Yost Page 46

prudential rule . . . dooms a plaintiff’s claims.” Henry v. Blank, 167 F.4th 375, 381 (6th Cir. 2026) (citing Kowalski, 543 U.S. at 134).

The reasoning undergirding Kowalski’s test supports this result. The general rule that a party “cannot rest his claim to relief on the legal rights or interests of third parties,” Kowalski, 543 U.S. at 129 (citation omitted), “‘frees the [c]ourt[s] not only from unnecessary pronouncement on constitutional issues, but also from premature interpretations of statutes in areas where their constitutional application might be cloudy,’ and it assures the court that the issues before it will be concrete and sharply presented,” Sec’y of State of Md. v. Joseph H. Munson Co., 467 U.S. 947, 955 (1984) (citations omitted). “Generally, the third party will be the best advocate of its own position, and the plaintiff [seeking standing under the third party’s rights] may place a slightly different, self-interested ‘spin’ on its presentation.” Amato v. Wilentz, 952 F.2d 742, 748 (3d Cir. 1991) (citing Munson, 467 U.S. at 955). These concerns are particularly pointed where, as here, the right-holding third parties are ostensibly benefited by the challenged statute. At the very least, large social-media operators and children cannot and do not share a “close relationship” with regard to a statute targeting (among other things) the operators’ terms of service.

NetChoice retorts that the Supreme Court has been more “forgiving” in certain circumstances: “‘Within the context of the First Amendment,’ for example, ‘the Court has enunciated other concerns that justify a lessening of prudential limitations on standing.’” Kowalski, 543 U.S. at 130 (quoting Munson, 467 U.S. at 956). Just so. But lessening does not mean universal elimination. And whereas similar or even dissimilar interests may clear a lessened close-relationship hurdle, the opposing interests at play here will not. To that point, the Supreme Court in Munson applied the two “prudential” prongs of the third-party-standing test with marked dissimilarity. See Munson, 467 U.S. at 956–60. In considering the Munson defendant’s lack-of-hindrance argument against third-party standing, the Court held that, in the context of an overbreadth challenge under the First Amendment, this failure was of no consequence. Id. at 957–58. “[W]here the claim is that a statute is overly broad in violation of the First Amendment, the Court has allowed a party to assert the rights of another without regard to the ability of the other to assert his own claims.” Id. at 957. But the Court considered the No. 25-3371 NetChoice, LLC v. Yost Page 47

merits of the close-relationship prong, finding that the defendant had “come forward with no reason why [the plaintiff] is an inadequate advocate to assert the [third parties’] rights. The activity sought to be protected [wa]s at the heart of the business relationship between [the plaintiff] and [its third-party clients], and [the plaintiff’s] interests in challenging the statute we[re] completely consistent with the First Amendment interests of the [third parties] it represents.” Id. at 958 (emphasis added). Having been satisfied that no close-relationship barrier existed, the Court “s[aw] no prudential reason not to allow [the plaintiff] to challenge the statute” based on third-party standing.” Id.

Even under the reduced burden applicable to overbreadth challenges, NetChoice’s interests vis-à-vis those of minors are not “completely consistent,” “closely aligned,” or even neutral. They are opposed to one another. Nor am I concerned that “there is a possibility that, rather than risk punishment for [their] conduct in challenging the statute, [third parties] will refrain from engaging further in the protected activity.” Munson, 467 U.S. at 956. This concern—the “danger of chilling free speech”—undergirding Munson’s and Virginia v. American Booksellers Association, Inc.’s relaxed third-party-standing standard is absent where, as here, the third parties do not “risk punishment” or some other harm for speaking freely. See id.; see also Am. Booksellers, 484 U.S. 383, 389, 392–93 (1988), certified question answered sub nom. Commonwealth v. Am. Booksellers Ass’n, Inc., 236 Va. 168 (1988) (permitting booksellers to assert the rights of book-buying adults reluctant to enter an adults-only store or section of a store “because of a judicial prediction or assumption that the statute’s very existence may cause others not before the court to refrain from constitutionally protected speech or expression” (quoting Munson, 467 U.S. at 956–57)). Perhaps some minor users face limited access to social- media platforms under The Act, but there is no risk that the state will punish them for their speech or that they will otherwise be “chilled” and “refrain from constitutionally protected speech.” The Act creates no cause of action against minors. And NetChoice does not contend that minors cannot bring their own challenges to The Act.

NetChoice’s counterarguments are unavailing. First, NetChoice contends this is an improper “inject[ion of] merits issues into the standing analysis.” Appellee NetChoice’s Br. at 27. But NetChoice does not explain how a court can consider NetChoice’s ability to “frame the No. 25-3371 NetChoice, LLC v. Yost Page 48

issues and present them with necessary adversarial zeal,” see Munson, 467 U.S. at 956, without acknowledging the adversarial issues. To that point, NetChoice itself engages the merits in its effort to prove its alignment with children’s interests. Appellee NetChoice’s Br. at 26–27. At any rate, a court may readily recognize a conflict of interests over a particular claim without prejudging the merits of the claim.

Second, NetChoice contends that Elk Grove Unified School District v. Newdow, 542 U.S. 1 (2004), which Yost cites in support of his argument that conflicts of interest defeat third-party standing, is inapposite. Appellee NetChoice’s Br. at 27. But Yost does not need Newdow to show that conflicting interests are likely fatal (and are certainly fatal in this case) to a finding of a close relationship. Munson, Kowalski, and the logic of adversarial proceedings make this self- evident point.

Third, NetChoice cannot rely on American Booksellers, 484 U.S. at 392–93. While American Booksellers affirmed Munson’s holding lowering the third-party-standing standard in overbreadth cases, the American Booksellers Court did not consider third-party-standing conflicts of interest. See id. Nor are any such issues present in that case. The plaintiffs (booksellers and booksellers’ associations) sought to enjoin enforcement of a statute restricting the open display of sexual content in bookstores, but the plaintiffs were suing on behalf of adult purchasers of such material. See id. at 386, 388 n.3, 389, 393. The booksellers did not—as would be analogous to NetChoice’s standing argument—assert minors’ rights to view the sexual material unhindered.

Fourth, NetChoice argues that third-party standing requires only an alignment of interests “insofar as it is linked to the right asserted.” Appellee NetChoice’s Br. at 27 (quoting Yost, 778 F. Supp. 3d at 944). But NetChoice identifies no controlling caselaw in support of this principle. Instead, NetChoice and the district court rely on a Third Circuit case: Amato, 952 F.2d at 752. Amato is not binding on this panel; nor does Amato help NetChoice’s case. While the Third Circuit did hold that “the relationship between the third party and the plaintiff only counts insofar as it is linked to the right asserted,” context reveals that this requirement was meant to protect the third party’s interests by limiting the kinds of close relationships the plaintiff may assert: No. 25-3371 NetChoice, LLC v. Yost Page 49

But a close personal relationship is neither necessary nor sufficient for third party standing. Even a close relative will not be heard to raise positions contrary to the interests of the third party whose rights he or she claims to represent: the litigant would then hardly be a vigorous advocate of the third party’s position. Moreover, the relationship between the third party and the plaintiff only counts insofar as it is linked to the right asserted, for it is when “the enjoyment of the right is inextricably bound up with the activity the litigant wishes to pursue [that] the court can at least be sure that its construction of the right is not unnecessary . . . .”

Id. at 751–52 (citations omitted) (emphasis added). The focus on the “right asserted” is a limitation on the litigant, not a permission slip to cherry-pick certain rights and render them devoid of all context in an effort to ignore relevant conflicting interests.

Finally, NetChoice relies on other courts’ treatment (explicit or tacit) of NetChoice’s third-party standing in related litigation. See Appellee NetChoice’s Br. at 25. These decisions are not binding on this panel. Nor are they convincing. Of the decisions that remain good law, they either do not address third-party standing, do not address conflicts of interests in the context of third-party standing, or overextend American Booksellers and related cases to dismiss such conflicts as irrelevant in overbreadth challenges—a position not supported by American Booksellers and contrary to the Supreme Court’s on-point treatment in Munson.

As I understand it, Judge Ritz would permit NetChoice’s reliance on minors’ standing, reasoning that “the question is not whether the interests of NetChoice’s members are aligned with all of their minor users’ interests, but whether NetChoice’s members are ‘fully, or very nearly, as effective a proponent of the right’ at issue as the minor users.” See post, Op. of J. Ritz, at 2 (emphasis in original) (quoting Singleton v. Wulff, 428 U.S. 106, 115 (1976) (plurality opinion)). Judge Ritz would hold that “the activity [NetChoice’s members] wish[] to pursue” is “inextricably bound up” with their potential minor users’ rights. Id. (quoting Singleton, 428 U.S. at 114 (plurality opinion)) (alterations added).

But I read Singleton differently. Justice Blackmun’s opinion acknowledges that where a nonparty’s “enjoyment of [a] right is inextricably bound up with the activity the litigant wishes to pursue, the court at least can be sure that its construction of the right is not unnecessary in the sense that the right’s enjoyment will be unaffected by the outcome of the suit.” 428 U.S. at 114– No. 25-3371 NetChoice, LLC v. Yost Page 50

15 (plurality opinion) (emphasis added). Justice Blackmun further observed that, in such cases, “the relationship between the litigant and the third party may be such that the former is fully, or very nearly, as effective a proponent of the right as the latter.” Id. at 115 (emphasis added) (plurality opinion). So, an “inextricably bound up” nonparty-right-to-litigant-activity relationship merely guarantees that the right will not be completely “unaffected” by the litigation, and such a relationship may secure effective third-party representation. In other words, even if NetChoice’s members’ desired activities were “inextricably bound up” with minors’ rights, that would be a necessary, but not necessarily sufficient, prerequisite to NetChoice’s asserting those rights.

To that point, Singleton’s third-party-standing analysis does not at all address situations where, as here, litigants and nonparties diverge in their relevant interests. And even if, in order to assert third-party standing, a litigant must be an effective proponent of only the particular rights “at issue,” see post, Op. of J. Ritz, at 2 (emphasis removed), that would not permit NetChoice to frame the at-issue rights so narrowly as to ignore obviously different interests of minors in how those rights are protected. Nor do I agree that NetChoice can brush aside these conflicts merely because its members’ relationships with minor users would be ongoing, and I am not convinced that we should read into Craig v. Boren, 429 U.S. 190 (1976), a consideration of conflicting interests that is absent from that opinion’s text. Contra post, Op. of J. Ritz, at 3.

Finally, Judge Ritz would permit NetChoice’s third-party standing because minors are hindered in asserting their own rights. See post, Op. of J. Ritz, at 4. I am not convinced that this is the case. For one, minors in Texas have already brought their own challenge (parallel to a challenge by a NetChoice-esque technology industry association) to a Texas statute similar to The Act. See Students Engaged in Advancing Tex. v. Paxton, 765 F. Supp. 3d 575, 582 (W.D. Tex. 2025). And I fail to see how minors’ purported right to speak anonymously on social media—a right which NetChoice does not discuss on appeal—is destroyed by minors vindicating their own purported rights to join social media platforms. Contra post, Op. of J. Ritz, at 4. In that vein, “swingers” interests in anonymously vindicating their purported right to advertise in a sexual magazine do not map onto the interests at issue here. Contra id. (discussing Connection Distrib. Co. v. Reno, 154 F.3d 281, 295 (6th Cir. 1998)). There is no evidence that any minor has No. 25-3371 NetChoice, LLC v. Yost Page 51

refrained from suing Yost for fear of revealing that he or she is a minor planning on generally speaking on a social media platform.

Judge Clay takes a slightly different approach to the third-party standing analysis, adopting the Third Circuit’s framing of the issue as a “balance of factors.” See Amato, 952 F.2d at 750. But the Third Circuit derives this balancing of factors from the Supreme Court’s third- party-standing analysis in Caplin & Drysdale, Chartered v. United States, 491 U.S. 617, 623 n.3 (1989). See Amato, 952 F.2d at 749. Caplin & Drysdale is a Sixth Amendment case, from which, as Amato acknowledges, the Supreme Court’s third-party-standing analysis in First Amendment cases differs dramatically. See id. (observing that Munson and American Booksellers discount Caplin & Drysdale’s obstacle factor entirely). In fact, the Supreme Court would go on to explicitly distinguish its First Amendment treatment of third-party standing from its analysis in Caplin & Drysdale. See Kowalski, 543 U.S. at 130–31. Admittedly, “[t]he limitations on third-party standing are hard[] to classify,” Lexmark, 572 U.S. at 127 n.3, and the test of those limitations lacks precise definition. But, in light of Kowalski, I would not employ Amato’s balancing approach in a First Amendment case like this one. Instead, I make only the narrow observation that third-party standing is almost certainly impermissible in any case where the plaintiff’s and the third parties’ interests are at odds. And it is certainly impermissible on the facts before us here. NetChoice may bring this lawsuit on behalf of its members, raising any free-speech rights those members possess in relation to The Act. But NetChoice may not raise the rights of minors.

NetChoice first challenges The Act under the First Amendment. Courts have taken diverse approaches to NetChoice’s First Amendment challenges to similar legislation. Some have applied strict scrutiny, see, e.g., Jones, 2026 WL 561099 at *10; others have applied intermediate scrutiny, see, e.g., Uthmeier, 2025 WL 3458571 at *3. The district court below held that Ohio’s Act implicates First Amendment review, warrants strict scrutiny, and fails under that test. Yost, 778 F. Supp. 3d at 947–57. Judges Clay and Ritz both hold that The Act regulates speech on the basis of content, so strict scrutiny applies. But whereas Judge Clay holds that The Act survives strict scrutiny, Judge Ritz disagrees. I am skeptical that The Act is a No. 25-3371 NetChoice, LLC v. Yost Page 52

content-based regulation in the way Judges Clay and Ritz say it is, so I question whether strict scrutiny (as opposed to intermediate scrutiny) bears on to those applications of The Act considered by my colleagues.

But a step remains before we analyze whether The Act regulates speech and what level of scrutiny applies to that regulation. NetChoice does not challenge The Act as applied to its members. Rather, it asserts only that the statute is facially invalid. “A facial challenge to a legislative Act is, of course, the most difficult challenge to mount successfully, since the challenger must establish that no set of circumstances exists under which the Act would be valid.” United States v. Salerno, 481 U.S. 739, 745 (1987). This exacting standard is lowered in First Amendment cases: “The question is whether ‘a substantial number of [the law’s] applications are unconstitutional, judged in relation to the statute’s plainly legitimate sweep.’” Moody, 603 U.S. at 723 (2024) (quoting Americans for Prosperity Found. v. Bonta, 594 U.S. 595, 615 (2021)). In other words, The Act’s “plainly legitimate sweep” may be struck down, but only if its “unconstitutional applications substantially outweigh its constitutional ones.” Id. at 723–24. “Even in free-speech cases, . . . facial invalidation of a statute remains ‘strong medicine that is not to be casually employed.’” Connection Distrib. Co. v. Holder, 557 F.3d 321, 336 (6th Cir. 2009) (quoting United States v. Williams, 553 U.S. 285, 293 (2008)).

In short, “NetChoice chose to litigate th[is] case[] as [a] facial challenge[], and that decision comes at a cost.” See Moody, 603 U.S. at 723. And just as happened in Moody, NetChoice failed to meet its burden here. The timeline of the proceedings below demonstrates this failure. “One week after the parties’ summary judgment briefing concluded, the Supreme Court decided Moody v. NetChoice, LLC,” in which the Supreme Court explained that “[a] proper First Amendment facial challenge . . . proceeds in two steps.” Yost, 778 F. Supp. 3d at 936. “The first step in the proper facial analysis is to assess the state law[’s] scope.” Moody, 603 U.S. at 724. “The next order of business is to decide which of the laws’ applications violate the First Amendment, and to measure them against the rest.” Id. at 725. After the Supreme Court’s decision in Moody, the circuit courts from which the cases in that consolidated appeal originated remanded their respective cases to the respective district courts; one explicitly noted that Moody raises “fact-intensive questions that must be answered by the district court in the first No. 25-3371 NetChoice, LLC v. Yost Page 53

instance after thorough discovery.” See Yost, 778 F. Supp. 3d at 936–37 (quoting NetChoice, L.L.C. v. Paxton, 121 F.4th 494, 499 (5th Cir. 2024)). In light of this, the district court in this case “asked the parties . . . whether discovery should be re-opened or summary judgment briefing should be supplemented.” Id. at 937. NetChoice, which is the party that bears the burden under Moody’s framework, Moody, 603 U.S. at 744, “maintained that the record is sufficient under Moody,” Yost, 778 F. Supp. 3d at 937. Yost “opined that the record is insufficient but that the burden of proof rests with NetChoice.” Id.

Unsurprisingly, NetChoice failed to satisfy its facial challenge burden in clairvoyant anticipation of Moody’s holding. NetChoice did not thoroughly “assess [The Act’s] scope” or provide an adequate record “to decide which of [The Act’s] applications violate the First Amendment, and to measure them against the rest.” See Moody, 603 U.S. at 724–25. The district court likewise did not attempt this inquiry, concluding that “unlike the provisions at issue in Moody, the regulation here ‘raises the same First Amendment issues’ ‘in every application to a covered business.’” Yost, 778 F. Supp. 3d at 947 (quoting NetChoice, LLC v. Bonta (Bonta I), 113 F.4th 1101, 1116 (9th Cir. 2024)).

Presumably, the district court made this assumption because it permitted NetChoice to assert the rights of its members’ potential minor users. Id. at 946. Minors’ rights to access social media platforms and receive and publish speech might present a more unitary First Amendment analysis. But I decline to decide here whether, relying on such third-party standing, NetChoice can demonstrate that The Act raises the same constitutional issues in every application. As addressed above, the district court’s third-party standing conclusion was in error, and NetChoice may assert only its members’ rights. And that posture does not permit NetChoice to shirk its burden under Moody. In fact, one need look no further than the Supreme Court’s analysis in that case to identify the problems with that shortcut. From the record below, I cannot assess the full scope of NetChoice’s members’ services covered by The Act, much less the scope of services offered by nonparties. Some of those services, such as “direct messaging service[s]” may not warrant the same kind of constitutional protection (at least with regard to the service provider’s right to speak freely) as other services, such as “[c]urating a [social-media] feed.” Moody, 603 U.S. at 725–26. In other words, “regulation of those diverse activities could well fall on different No. 25-3371 NetChoice, LLC v. Yost Page 54

sides of the constitutional line.” Id. at 726. “[C]onstitutional ‘answers might differ’ depending on the speaker at issue,” “the platforms at issue,” and “the speech at issue.” Doe v. Burlew, 165 F.4th 525, 534–35 (6th Cir. 2026) (quoting Moody, 603 U.S. at 725.)

In sum, this court cannot “explore the [Act’s] full range of applications—the constitutionally impermissible and permissible both—and compare the two sets.” See Moody, 603 U.S. at 726. Rather, it is left in the same conundrum as in Moody: “[W]e are a court of review, not of first view.” Id. (citation omitted). But unlike the procedural situation in Moody and Doe, NetChoice was fully aware of its burden: the Supreme Court’s decision in Moody was released prior to the summary-judgment hearing in this case, and the district court appropriately questioned NetChoice whether the record was complete in light of the decision. NetChoice—the same party that litigated the Moody case—answered in the affirmative. Having declined the opportunity to seek leave to file additional summary-judgment briefing in light of Moody or request a reopening of discovery, NetChoice must pay the “cost” of its decision to bring a facial challenge without a sufficiently developed record. See Moody, 603 U.S. at 723.

Judge Clay reaches the merits of NetChoice’s First Amendment challenge because he holds that The Act “burdens at least some protected speech.” I do not disagree that The Act almost certainly burdens some speech, and some of that speech may very well be protected by the First Amendment. But I read Moody as requiring a plaintiff bringing a facial First Amendment challenge to identify the full scope of that speech and then assess how much of that burden is unjustified in light of the First Amendment. To engage in the merits of NetChoice’s challenge without its having met this facial burden would likely involve our analyzing only the “heartland applications” of The Act and attempting to extrapolate the constitutionality of those applications to The Act’s entire scope—an approach forbidden by Moody, 603 U.S. at 724. So while some—perhaps even most—of The Act’s applications warrant constitutional review, I refrain from doing so on a facial challenge without a developed record.

Judge Ritz also says that we may reach the merits of NetChoice’s challenge; he adopts the district court’s reasoning that because NetChoice’s members and other covered operators are “under the same statutory obligation to block access to unemancipated Ohio minors absent parental consent,” The Act “raises the same First Amendment issues in every application to a No. 25-3371 NetChoice, LLC v. Yost Page 55

covered business.” Yost, 778 F. Supp. 3d at 947 (citation modified). As discussed above, this logic relies on NetChoice’s assertion of minors’ rights. Because I, along with Judge Clay, do not think that third-party standing is proper, I would analyze NetChoice’s facial challenge burden in light of its members’ First Amendment rights alone. And doing so does not turn NetChoice’s facial claim into an as-applied challenge. See contra post, Op. of J. Ritz, at 6. NetChoice’s members may challenge The Act as unconstitutionally enforced against them “in any circumstance,” as opposed to unconstitutional as applied to those members’ “specific conduct.” See Doe, 165 F.4th at 530 (emphasis removed). Nor does denial of third-party standing as to minors necessarily prevent NetChoice from asserting a facial challenge with regard to other operators, though NetChoice does not appear to raise that argument on appeal. At any rate, NetChoice’s decision to bring a facial challenge based on third-party interests that it may not assert does not require us to disregard our third-party-standing doctrines in order to save that facial challenge. And because The Act contemplates applications with potentially disparate First Amendment implications (i.e., hosting curated social-media feeds versus facilitating peer-to-peer messaging, see Ohio Rev. Code § 1349.09(A)(1)(d)), NetChoice may not assert its own rights in a one-size-fits-all facial challenge.

C. NetChoice’s Vaguness Challenge

NetChoice also contends that The Act is unconstitutionally vague on its face based on two provisions in particular: (1) a coverage provision limiting The Act’s application to operators of “online web site[s], service[s], or product[s] that target[] children, or is reasonably anticipated to be accessed by children,” Ohio Rev. Code § 1349.09(B), and (2) an exclusion from coverage of “online web site[s], service[s], or product[s] respecting which interaction between users is limited to . . . [c]omments incidental to content posted by an established and widely recognized media outlet, the primary purpose of which is to report news and current events.” Id. § 1349.09(O)(2).

Under the Supreme Court’s test for vagueness, NetChoice “must show either that the [Act’s] provisions (1) ‘fail to provide people of ordinary intelligence a reasonable opportunity to understand what conduct it prohibits’ or (2) ‘authorize or even encourage arbitrary and discriminatory enforcement.’” Platt v. Bd. of Comm’rs on Grievances & Discipline of Ohio Sup. No. 25-3371 NetChoice, LLC v. Yost Page 56

Ct., 894 F.3d 235, 246 (6th Cir. 2018) (citation modified) (quoting Hill v. Colorado, 530 U.S. 703, 732 (2000)). “While ‘a more stringent vagueness test should apply’ to laws abridging the freedom of speech, that standard is relaxed somewhat when the [statute] imposes civil rather than criminal penalties and includes an implicit scienter requirement.” Id. (quoting Vill. of Hoffman Estates v. Flipside, Hoffman Estates, Inc., 455 U.S. 489, 499 (1982)). “An enactment imposing criminal sanctions or reaching a substantial amount of constitutionally protected conduct may withstand facial constitutional scrutiny only if it incorporates a high level of definiteness.” Belle Maer Harbor v. Charter Twp. of Harrison, 170 F.3d 553, 557 (6th Cir. 1999) (citing Hoffman Estates, 455 U.S. at 494). “But perfect clarity and precise guidance have never been required even of regulations that restrict expressive activity.” United States v. Williams, 553 U.S. 285, 304 (2008) (quoting Ward v. Rock Against Racism, 491 U.S. 781, 794 (1989)).

NetChoice’s failure to present the full scope of The Act and analyze the constitutional implications for every application rears its head in the facial-vagueness analysis as well. Without a more developed record, we are left to guess whether The Act truly covers significant amounts of protected speech. Arguably, Moody would also counsel us to reverse on NetChoice’s vagueness claim for failure to develop the factual record, similarly to Moody’s instruction with regard to First Amendment overbreadth claims. But Moody’s relationship with the void-for- vagueness doctrine is unclear. At any rate, because NetChoice’s vagueness claim fails even if The Act covers a significant amount of protected speech, I presume as much without deciding. So, mindful that The Act presumably covers significant amounts of protected speech and contains no scienter requirement but does not impose criminal punishment, I turn to the two purportedly vague provisions.

1. The Coverage Provision

First, NetChoice asserts that the coverage provision is vague, focusing primarily on The Act’s list of discretionary factors that the Ohio attorney general or a court “may consider” when “determining whether an operator’s online web site, service, or product targets children, or is reasonably anticipated to be accessed by children.” Ohio Rev. Code § 1349.09(C). According to NetChoice, “[a]lmost all these factors are vague, subjective, or so broad as to be meaningless,” and covered websites are at a loss as to “how [they] are supposed to weigh the ‘eleven No. 25-3371 NetChoice, LLC v. Yost Page 57

discretionary factors.’” Appellee NetChoice’s Br. at 72. But this misses the point. Even taking NetChoice’s characterization as true, these non-exhaustive factors are only in service of a clearly articulated standard: whether a website “targets children” or “is reasonably anticipated to be accessed by children.” Whether the list itself is vague or unbounded is irrelevant if the provision it seeks to exemplify is sufficiently clear. See Platt, 894 F.3d at 247 (holding that “[w]hen the common meaning of a word provides adequate notice of the prohibited conduct, the statute’s failure to define the term will not render the statute void for vagueness” (citation omitted)).

And while the district court admonished—without analysis and citing to no authority— that the statute’s focus on websites targeting or likely accessed by children amounts to facially “expansive language” that left the court befuddled and “would leave many operators unsure as to whether [The Act] applies to their website,” Yost, 778 F. Supp. 3d at 957, I cannot agree. Whether a website targets or is reasonably anticipated to be accessed by children does not “fail to provide people of ordinary intelligence a reasonable opportunity to understand” the statute’s scope. See Platt, 894 F.3d at 246 (citation modified). “The fact that a law is ‘marked by “flexibility and reasonable breadth, rather than meticulous specificity”’ does not render it unduly vague.” Id. at 246–47 (quoting Grayned v. City of Rockford, 408 U.S. 104, 110 (1972)). “Said another way, where the challenged language ‘is commonly used in both legal and common parlance,’ it often will be ‘sufficiently clear so that a reasonable person can understand its meaning.’” Id. at 247 (citation omitted). The Act’s coverage provision employs the kind of language used in legal and common parlance to articulate a comprehensible point: The Act is aimed at the social-media activities of children, not adults.

Relatedly, this court has rejected similar vagueness challenges to language defining the scope of a prohibition with regard to children. In United States v. Arnold, we held, albeit under a plain error standard, that a sex offender’s condition of release barring him from “association with children younger than 18 years of age” was not unconstitutionally vague. 549 F. App’x 491, 497–98 (6th Cir. 2013); see also United States v. Zobel, 696 F.3d 558, 575 (6th Cir. 2012) (similar). Our sister circuits have held that similar language is not vague in this context. See, e.g., United States v. Paul, 274 F.3d 155, 165–67 (5th Cir. 2001) (rejecting vagueness challenge to a condition of release instructing the releasee “to ‘avoid places, establishments, and areas No. 25-3371 NetChoice, LLC v. Yost Page 58

frequented by minors’”); United States v. MacMillen, 544 F.3d 71, 76 (2d Cir. 2008) (similar); United States v. Loy, 237 F.3d 251, 268–69 (3d Cir. 2001) (similar); United States v. Ristine, 335 F.3d 692, 696–97 (8th Cir. 2003) (similar); United States v. Mike, 632 F.3d 686, 697 (10th Cir. 2011) (similar); United States v. Gibson, 998 F.3d 415, 419 (9th Cir. 2021) (rejecting similar vagueness challenge to a condition including the phrase, “primarily used by children”). While these cases do not present an exact comparison, I see no reason why a statute defining its scope with regard to how a product “targets children” or “is reasonably anticipated to be accessed by children” is insufficiently articulate.

NetChoice offers little by way of rebuttal, and its argument is notably sparse on supporting authority. As mentioned above, its contention that the coverage provision fails for vagueness rests almost entirely on the purported vagueness of “eleven discretionary factors” supporting that provision. The only attack lobbed at the coverage provision itself is NetChoice’s concern that it fails “to distinguish among websites that target or can reasonably anticipate being accessed by 15-year-olds versus 16-year-olds.” Appellee NetChoice’s Br. at 73; see also NetChoice’s Mot. Summ. J., R. 43, PageID 661. But The Act does not require such hair splitting. Rather, The Act only requires covered operators to determine whether their websites target or can reasonably anticipate being accessed by minors under the age of sixteen, regardless of whether the same could be said about minors aged sixteen-years-and-one-day or older. And NetChoice’s remaining argument—its assertion that the coverage provision lends itself to arbitrary enforcement—is just conclusory. Again, whether the purportedly vague nature of the “eleven discretionary factors” risks arbitrary enforcement does not mean that the coverage provision itself bears such risk. And without its attacks on the list of factors, NetChoice asserts only that the coverage provision “grants [Yost] far too much discretion.” This will not do. Meriwether v. Hartop, 992 F.3d 492, 518 (6th Cir. 2021) (rejecting a similar “conclusory assertion” that an allegedly vague policy “g[ave] officials ‘unbridled discretion’ in enforcement”).

Finally, NetChoice contends that “[a] law that implicates First Amendment freedoms is unconstitutionally vague if it ‘reaches a substantial amount of constitutionally protected conduct.’” Appellee NetChoice’s Br. at 71 (quoting Belle Maer Harbor, 170 F.3d at 557). This No. 25-3371 NetChoice, LLC v. Yost Page 59

is dead wrong. In Belle Maer Harbor, we held that, “[i]n examining a facial challenge, this court must first ‘determine whether the enactment reaches a substantial amount of constitutionally protected conduct.’” Id., 170 F.3d at 557 (emphasis added) (quoting Hoffman Estates, 455 U.S. at 494). If it does, we do not automatically deem it unconstitutionally vague but inquire whether it “incorporates a high level of definiteness.” Id. (citing Hoffman Estates, 455 U.S. at 494). And while the Supreme Court has “permit[ed] plaintiffs to argue that a statute is overbroad because it is unclear whether it regulates a substantial amount of protected speech,” Williams, 553 U.S. at 304 (citation omitted) (emphasis added), it does not empower NetChoice to shoehorn a First Amendment overbreadth claim into a Fourteenth Amendment vagueness challenge when NetChoice has manifestly failed to meet its overbreadth burden under Moody. See supra, II.B; see also Holder v. Humanitarian L. Project, 561 U.S. 1, 20 (2010). Plaintiffs often bring First Amendment overbreadth and Fifth Amendment vagueness challenges together, but these doctrines attack a purportedly unconstitutional statute on different grounds; substituting the analysis for one challenge for the test for the other would render both “substantially redundant.” See Humanitarian L. Project, 561 U.S. at 20. I acknowledge that the coverage provision likely covers a substantial number of website operators, and this appears to be NetChoice’s chief complaint—even in its void-for-vagueness arguments. But a statute is not vague just because it has a wide berth. See id. (stating that Supreme Court “precedents make clear that a Fifth Amendment vagueness challenge does not turn on whether a law applies to a substantial amount of protected expression”). The proper vehicle to challenge the wide, but not vague, coverage of The Act is a challenge under the First Amendment.

2. The Media Exclusion

NetChoice also challenges as vague The Act’s exclusion of “online web site[s], service[s], or product[s] respecting which interaction between users is limited to . . . [c]omments incidental to content posted by an established and widely recognized media outlet, the primary purpose of which is to report news and current events.” Ohio Rev. Code § 1349.09(O)(2). Vagueness “analysis must turn on the ‘particular facts at issue, for a plaintiff who engages in some conduct that is clearly proscribed cannot complain of the vagueness of the law as applied to the conduct of others.’” Meriwether, 992 F.3d at 518 (quoting Humanitarian Law Project, 561 No. 25-3371 NetChoice, LLC v. Yost Page 60

U.S. at 18–19 (citation modified)). NetChoice does not argue that the media exclusion is vague as applied to any of its members. And the district court merely held that the media exclusion was “eyebrow-raising” because it “practically invites arbitrary application of the law,” again with no reference to whether it could be arbitrarily applied to NetChoice’s members. Nor is such a conclusion obvious: the media exclusion applies only to websites where user-to-user interaction is limited to comments incidental to news content. Many, if not all, of NetChoice’s members presumably permit user-to-user interaction relating to much more than news and current events. Since NetChoice does not claim that the media exclusion is vague as applied to its members, NetChoice’s remaining vagueness claim must also fail.2

III.

The Supreme Court has “made facial challenges hard to win.” Moody, 603 U.S. at 723. NetChoice did not satisfy its burden to facially challenge The Act. Nor did it satisfy its burden on its void-for-vagueness doctrine claim. For the foregoing reasons, I join in the judgment of this court to REVERSE the district court’s judgment and REMAND with instruction to enter judgment in favor of Yost.3

2 While I agree with some of Judge Clay’s analysis of NetChoice’s vagueness challenge, he ultimately reverses based on our holding in Libertarian Party of Ohio v. Husted that “a statute will be struck down as facially vague only if the plaintiff has demonstrated that the law is impermissibly vague in all of its applications.” 751 F.3d 403, 422 (2014) (citation modified). I am not sure that this part of Husted remains good law in light of Johnson v. United States, 576 U.S. 591, 603 (2015) (disagreeing with the dissent’s objection that “a statute is void for vagueness only if it is vague in all its applications”). Perhaps Husted states the correct standard, or perhaps a statute should be struck down for vagueness only when its unconstitutional applications substantially outweigh its constitutional ones, similar to the standard for a facial First Amendment challenge. But because NetChoice’s vagueness challenge fails on other grounds, we need not delve into this murky precedent. 3 Yost also contends on appeal that the district court’s injunctive relief was overbroad, see Trump v. CASA, Inc., 606 U.S. 831 (2025); Free Enterprise Fund v. Public Company Accounting Oversight Board, 561 U.S. 477 (2010). Because we vacate the injunction altogether, we need not decide whether that relief was overbroad. No. 25-3371 NetChoice, LLC v. Yost Page 61

DISSENT _________________

RITZ, Circuit Judge, dissenting. I see this challenging case differently than my colleagues do. As an initial matter, I would find that NetChoice has standing to assert the rights of its members’ minor users. And on the merits, I agree with Judge Clay that Ohio’s Parental Notification by Social Media Operators Act burdens speech protected by the First Amendment. I also agree with Judge Clay that the Act is content-based and thus subject to strict scrutiny. But I would hold, however, that the Act fails strict scrutiny.

So, I would affirm the district court’s judgment, although I would narrow the injunctive relief to apply only to NetChoice’s members. For these reasons, I respectfully dissent.

The parties agree that NetChoice has associational standing to bring a First Amendment challenge on behalf of its members. In my view, NetChoice also has standing to assert the rights of its members’ minor users.

Generally, a plaintiff may not assert the rights of third parties unless the plaintiff can establish “that it has suffered a concrete, redressable injury, that it has a close relation with the third party, and that there exists some hindrance to the third party’s ability to protect his or her own interests.” Connection Distrib. Co. v. Reno, 154 F.3d 281, 295 (6th Cir. 1998). The Supreme Court has been “quite forgiving” when applying this test to First Amendment challenges. Kowalski v. Tesmer, 543 U.S. 125, 130 (2004). This forgiving approach is rooted in “a judicial prediction or assumption that [a] statute’s very existence may cause others not before the court to refrain from constitutionally protected speech or expression.” Virginia v. Am. Booksellers Ass’n, Inc., 484 U.S. 383, 392-93 (1988) (citation modified).

To determine whether there is a close enough relationship for a litigant to bring suit on behalf of a third party, we ask whether “the relationship between the litigant and the third party may be such that the former is fully, or very nearly, as effective a proponent of the right as the No. 25-3371 NetChoice, LLC v. Yost Page 62

latter.” Powers v. Ohio, 499 U.S. 400, 413 (1991) (quoting Singleton v. Wulff, 428 U.S. 106, 115 (1976)). So here, the question is not whether the interests of NetChoice’s members are aligned with all of their minor users’ interests, but whether NetChoice’s members are “fully, or very nearly, as effective a proponent of the right” at issue as the minor users. Singleton, 428 U.S. at 115. This close-relationship standard is typically satisfied if the nonparty’s rights are “inextricably bound up with the activity the litigant wishes to pursue.” Id. at 114. For instance, “vendors and those in like positions have been uniformly permitted to resist efforts at restricting their operations by acting as advocates of the rights of third parties who seek access to their market or function.” Craig v. Boren, 429 U.S. 190, 195 (1976).

NetChoice meets this requirement. As Judge Clay correctly recognizes, NetChoice’s members “engage in their own protected expression by curating third-party speech.” Clay Op. at 18. And by restricting the First Amendment rights of NetChoice’s minor users, the Act restricts the speech that these users can curate and publish on their platforms. NetChoice and its members are incentivized to advocate for more expansive First Amendment rights for its users, because the members’ rights are tied to their users’ ability to express themselves. Put another way, the “enjoyment” of the users’ First Amendment rights are “inextricably bound up” with NetChoice’s members. Singleton, 428 U.S. at 114.

My colleagues discuss whether the minor users’ “best interests” align with NetChoice’s members generally, Clay Op. at 12, and whether NetChoice’s interests and the interests of minor users diverge, Batchelder Op. at 45-47. But again, “the relationship between the third party and the plaintiff only counts insofar as it is linked to the right asserted.” Amato v. Wilentz, 952 F.2d 742, 752 (3d Cir. 1991) (emphasis added). And here, the First Amendment interests of NetChoice overlap neatly with the First Amendment interests of its members’ minor users. Amato provides a useful counterexample. There, the New Jersey Supreme Court Chief Justice blocked Warner Brothers from using a state courthouse in Essex County as a film set. 952 F.2d at 744. Warner Brothers offered a $250,000 donation to the Courthouse Restoration Fund to secure filming rights but ultimately selected a different location. Id. at 744-45. Essex County then sued on behalf of Warner Brothers and alleged the Chief Justice’s decision caused the county to lose $250,000 in revenue. Id. at 743. The Third Circuit declined to allow Essex No. 25-3371 NetChoice, LLC v. Yost Page 63

County to assert prudential standing on behalf of the studio. Although Warner Brothers might have faced “only minor obstacles” if it wanted to bring its own suit, the studio could very well have decided not to sue out of a fear the state could suspend filming for all movies at all courthouses going forward (a consequence that came to pass). Id. at 755. Essex County did not, however, share this broad fear or concern. Rather, the county merely wanted to recoup the money it lost during the “singular” transaction at issue. Id. at 753. So it was not clear that the parties’ “views coincided at the time suit was brought.” Id.

Here, NetChoice faces no comparable conflicts in vindicating the First Amendment rights of its members’ users, and the nature of their relationship is not based on a singular transaction but rather ongoing. So Amato is inapposite. I would instead look to the more analogous (and binding) cases of American Booksellers and Craig. In American Booksellers, the Supreme Court held that a bookstore could make First Amendment arguments on behalf of its customers. 484 U.S. at 392-93. And in Craig, the Court found that an alcohol vendor had standing to bring an equal-protection challenge to a state law that prohibited alcohol purchases for female customers under 18 years old and male customers under 21 years old. 429 U.S. at 192. Even though the alcohol vendor had a profit motive to sell alcohol—and consumption of alcohol by young men had well-documented adverse effects—the vendor could adequately represent the rights of male customers covered by the statute. Id. 195-97. Here, as in American Booksellers and Craig, “it is precisely NetChoice members’ financial incentive to keep minors on their platforms that suggests that they are fully, or very nearly, as effective a proponent of the minors’ rights to access the platforms.” NetChoice, LLC v. Yost, 778 F. Supp. 3d 923, 945 (S.D. Ohio 2025) (citation omitted).

Additionally, NetChoice’s members’ minor users are hindered from asserting their own rights. The minor users are not directly subject to the Act, so they are “denied a forum in which to assert their own rights.” Eisenstadt v. Baird, 405 U.S. 438, 446 (1972). Also, one aspect of the First Amendment interests at stake is the ability of minor users to “remain anonymous in their speech.” Connection Distrib., 154 F.3d at 295; see also McIntyre v. Ohio Elections Comm’n, 514 U.S. 334, 342 (1995) (noting anonymous expression is “an aspect of the freedom of speech protected by the First Amendment”). “[T]o require that the right be claimed by the No. 25-3371 NetChoice, LLC v. Yost Page 64

[users] themselves would result in nullification of the right at the very moment of its assertion.” Connection Distrib., 154 F.3d at 295 (citation modified). Finally, unlike the well-funded and sophisticated parties in Amato, minor users face “daunting” barriers to suit, considering “the small financial stake involved and the economic burdens of litigation.” Powers, 499 U.S. at 414- 15.

For these reasons, NetChoice is well-positioned to intervene on behalf of its member platforms’ minor users. I would therefore join the many other courts that have found NetChoice has prudential standing to assert its member users’ First Amendment rights in the face of similar state laws. See, e.g., NetChoice, L.L.C. v. Fitch, 134 F.4th 799, 805-07 (5th Cir. 2025); NetChoice v. Jones, 822 F. Supp. 3d 656, 668-69 (E.D. Va. 2026), appeal docketed, No. 26-1252 (4th Cir. Mar. 6, 2026); NetChoice v. Murrill, 812 F. Supp. 3d 594, 632 (M.D. La. 2025); NetChoice v. Carr, 789 F. Supp. 3d 1200, 1215-16 (N.D. Ga. 2025); Comput. & Commc’ns Indus. Ass’n v. Uthmeier, — F. Supp. 3d —, No. 4:24cv438-MW/MAF, 2025 WL 1570007, at *9 (N.D. Fla. June 3, 2025); NetChoice, LLC v. Griffin, No. 5:23-CV-05105, 2023 WL 5660155, at *12 (W.D. Ark. Aug. 31, 2023).

NetChoice brings a facial First Amendment challenge to the Act. “[T]he Supreme Court’s ‘overbreadth’ doctrine reduces a free-speech challenger’s burden to succeed on a facial claim as compared to the burden on a plaintiff invoking a different right.” Doe v. Burlew, 165 F.4th 525, 532 (6th Cir. 2026). A free-speech challenger can succeed on a facial overbreadth theory by showing that a law “prohibits a substantial amount of protected speech as compared to its plainly legitimate sweep.” Id. at 530 (citation omitted). Under Moody v. NetChoice, 603 U.S. 707 (2024), a court assessing an overbreadth challenge “must evaluate a law’s ‘scope’ by considering the ‘activities’ it covers and the ‘actors’ it regulates,” Doe, 165 F.4th at 532 (quoting 603 U.S. at 724), and then “decide which of the law[’s] applications violate the First Amendment, and . . . measure them against the rest.” Moody, 603 U.S. at 725.

Here, the scope of the Act is “largely not in dispute.” NetChoice, LLC v. Griffin, No. 5:23-CV-5105, 2025 WL 978607, at *7 (W.D. Ark. Mar. 31, 2025). Unlike the lower courts in No. 25-3371 NetChoice, LLC v. Yost Page 65

Moody, the district court examined the Act’s scope and weighed its constitutional and unconstitutional applications. Yost, 778 F. Supp. 3d at 947-57; cf. Moody, 603 U.S. at 726 (“[W]e are a court of review, not of first view.”) (citation omitted). Both parties introduced voluminous materials into the district court record as they litigated a temporary restraining order, a preliminary injunction, and cross-motions for summary judgment. After the Supreme Court issued Moody, the district court consulted with the parties and ultimately decided, based on the existing record, to grant summary judgment and a permanent injunction to NetChoice.

I would affirm this decision and, unlike Judge Batchelder, I see no reason to remand for further consideration of how Moody affects this case. Because this is a facial challenge, we must consider how the law applies not just to NetChoice’s members, but also its members’ users. To do otherwise would turn NetChoice’s claims into as-applied challenges. See NetChoice, LLC v. Bonta, 170 F.4th 744, 759 (9th Cir. 2026) (“[T]he focus on whether and how these provisions may impact social media companies, without considering any other potential applications, treats NetChoice’s challenges more like as-applied claims than like facial ones.”) (citation modified). And as explained above, the parental-consent provision at issue imposes a similar burden on not just NetChoice’s members, but also its member platforms’ users, across all the platforms regulated by the Act. As the district court put it, unlike the law at issue in Moody, “in every application to a covered website, the Act raises the same First Amendment issues,” Yost, 778 F. Supp. 3d at 947, and we can conduct the strict-scrutiny analysis “without reference to the intricacies of each individual platform’s operation,” Uthmeier, 2025 WL 1570007, at *9.

The Act in this case is likewise different from the statute at issue in Doe, which concerned a law that required registered sex offenders to display their full legal name on their social media accounts. Doe, 165 F.4th at 527. There, we noted that although some applications of the statute could raise constitutional concerns, such as when the regulated individuals sought to engage in anonymous political speech, other applications did not, such as when those individuals sought to message children anonymously on “children-focused” websites. See id. at 533-35. Here, in contrast, the platforms are all “under the same statutory obligation to block access to unemancipated Ohio minors absent parental consent,” so the Act “raises the same First Amendment issues in every application to a covered business.” Yost, 778 F. Supp. 3d at 947. No. 25-3371 NetChoice, LLC v. Yost Page 66

In response to the district court’s persuasive analysis of the Act’s unconstitutional footprint, the state has not explained how the Act could constitutionally be applied to the platforms the state seeks to regulate in a way that would not infringe on minor users’ rights. Notably, unlike the statute at issue in Moody, the Act only regulates websites that allow users to “[i]nteract socially with other users within the confines of the [website]” and “[c]reate or post content viewable by others.” Ohio Rev. Code § 1349.09(A)(1)(a), (d). The Act, therefore, exempts applications that might “fall on different sides of the constitutional line,” such as “email,” “online marketplace[s],” “payment service[s],” and “ride-sharing service[s].” Moody, 603 U.S. at 718, 725-26. So there is no need to sort through whether applying the Act to such sites would be constitutional; it does not apply to them.

Finally, even if the state could identify discrete constitutional applications of the parental- consent provision, for the reasons explained below NetChoice has established that the law’s unconstitutional applications are “substantially disproportionate to the statute’s lawful sweep.” See Doe, 165 F.4th at 532. In sum, like Judge Clay and the district court, I believe we can analyze NetChoice’s facial challenge on the existing record.

The crux of this case is whether the Act’s parental-consent provision violates the First Amendment. For the reasons ably set out by Judge Clay and the district court, as well as many other courts that have analyzed similar laws, I agree that the Act burdens speech and is content- based, so strict scrutiny applies. Clay Op. at 17-21; Yost, 778 F. Supp. 3d at 948-54; Carr, 789 F. Supp. 3d at 1218-27; Griffin, 2025 WL 978607, at *8-10; NetChoice, LLC v. Reyes, 748 F. Supp. 3d 1105, 1121 (D. Utah 2024); Murrill, 812 F. Supp. 3d at 643-47; see also Computer & Commc’ns Indus. Ass’n v. Uthmeier, No. 25-11881, 2025 WL 3458571, at *14-17 (11th Cir. Nov. 25, 2025) (Rosenbaum, J., dissenting).

But like the district court and many of these courts, I would hold that the Act fails to survive strict scrutiny. See also NetChoice, LLC v. Fitch, 145 S. Ct. 2658 (2025) (Kavanaugh, J., concurring in the denial of the application to vacate stay) (stating that enforcement of a similar Mississippi law “would likely violate [NetChoice’s] First Amendment rights under this Court’s No. 25-3371 NetChoice, LLC v. Yost Page 67

precedents”). Strict scrutiny is “the most demanding test known to constitutional law.” City of Boerne v. Flores, 521 U.S. 507, 534 (1997). To survive this test, the government must show its law is “narrowly tailored to serve compelling state interests.” KenAmerican Res., Inc. v. United States Sec’y of Lab., 33 F.4th 884, 893 (6th Cir. 2022) (citation omitted). In particular, it must “specifically identify an actual problem in need of solving” and show that “the curtailment of free speech” is “actually necessary to the solution.” Brown v. Ent. Merchants Ass’n, 564 U.S. 786,799 (2011) (citation omitted). “It is rare that a regulation restricting speech because of its content will ever be permissible.” United States v. Playboy Ent. Grp., Inc., 529 U.S. 803, 818 (2000).

A.

Yost cannot meet this high bar. At the outset, it is worth emphasizing that “[m]inors are entitled to a significant measure of First Amendment protection, and only in relatively narrow and well-defined circumstances may government bar public dissemination of protected materials to them.” Brown, 564 U.S. at 794 (citation modified). After all, “[a] fundamental principle of the First Amendment is that all persons have access to places where they can speak and listen, and then, after reflection, speak and listen once more.” Packingham v. North Carolina, 582 U.S. 98, 104 (2017). Although the websites the Act seeks to regulate may pose serious risks to children, these sites are also zones for constitutionally protected speech. And they cultivate “vast democratic forums” with the “potential to alter how we think, express ourselves, and define who we want to be.” Id. at 104-05 (citation omitted). Indeed, the Supreme Court has described social media platforms as “perhaps the most powerful mechanisms available to a private citizen to make his or her voice heard.” Id. at 107.

The parental-consent provision at the core of the Act represents a significant burden on the rights of NetChoice’s minor users to avail themselves of these “powerful mechanisms” of speech. Id. For many minor users, in fact, the Act would represent an “insurmountable [ ] barrier to entry for online speech.” Carr, 789 F. Supp. 3d at 1223; see also Uthmeier, 2025 WL 3458571, at *17 (Rosenbaum, J., dissenting) (“When it comes to speaking online, the Act effectively prohibits many minors from speaking at all.”). Like my colleagues and Yost, I do not doubt “that unfettered social media access can and does harm minors.” Griffin, 2025 WL No. 25-3371 NetChoice, LLC v. Yost Page 68

978607, at * 11. But the state does not have “a free-floating power to restrict the ideas to which children may be exposed.” Brown, 564 U.S. at 794. “We should be wary of governments supplanting parents in deciding which ideas children should and should not be exposed to.” Uthmeier, 2025 WL 3458571, at *18 (Rosenbaum, J, dissenting). Besides, the question under strict scrutiny “is not whether [the Act] would be effective, but whether there are less restrictive alternatives.” Jones, 822 F. Supp. 3d at 676.

B.

Against this backdrop, Yost argues that the Act furthers the governmental interests of protecting children, involving parents, and regulating contracts. But the Act’s provisions are both under- and over-inclusive and not “actually necessary” to further the state’s interests. Brown, 564 U.S. at 799. The Act thus fails strict scrutiny.

1.

First, it is true that protecting the “physical and psychological well-being of minors” is a compelling state interest. Sable Commc’ns of Calif., Inc. v. F.C.C., 492 U.S. 115, 126 (1989). But the Act is not narrowly tailored to support this interest.

Brown is instructive. There, the Supreme Court considered a law that prohibited selling or renting violent video games to minors, though minors could still access these games if purchased by an adult. Brown, 564 U.S. at 789, 802. In enacting this law, the legislature created a “wholly new category of content-based regulation that [wa]s permissible only for speech directed at children.” Id. at 794. The Supreme Court “emphatically” rejected this law as “unprecedented and mistaken.” Id. at 792, 794. And although the Court recognized that the legislature’s goal of “addressing a serious social problem” was “legitimate,” the solution was both “seriously underinclusive” and “seriously overinclusive.” Id. at 805.

Ohio has made the same error. Under the Act, parents need only give one-time consent for their children to create an account on covered websites, after which children can continue using the websites without any additional oversight. Neither the state nor Judge Clay’s opinion explains how this regime will prevent the websites from causing “issues with sleep, anxiety, No. 25-3371 NetChoice, LLC v. Yost Page 69

body dysmorphia, depression, and bullying” or “physiological effects like those from substance and gambling addictions.” Clay Op. at 3. Like Brown, if the sites the Act aims to cover are as “dangerous [and] mind-altering” as the government alleges, then allowing children to access these sites “so long as one parent . . . says it’s OK” is “wildly underinclusive.” Brown, 564 U.S. at 802. True, “the Act need not be perfectly drawn,” Clay Op. at 23 (citation modified), but it is unclear how one-time parental consent is “actually necessary to the solution,” Brown, 566 U.S. at 799. Judge Clay argues the “supporting evidence in Brown” as to the negative effects of violent video games “was weaker than Yost’s evidence.” Clay Op. at 22. But even so, the Supreme Court assumed in Brown that children’s access to violent video games was “a serious social problem” and nonetheless concluded that one-time parental consent was not a narrowly tailored solution. Brown, 564 U.S. at 799. The same analysis should govern here.

The Act’s efforts to minimize harm to children are also over-inclusive. As the district court aptly noted, the Act is a “breathtakingly blunt instrument for reducing social media’s harm to children.” Yost, 778 F. Supp. 3d at 956. To illustrate, the Act restricts minors’ access to a category of websites, even if those websites do not create the harms that the Act was written to address. “Ohio’s response to a societal worry that children might be harmed if they are allowed to access adult-only sections cannot be to ban children from the library altogether absent a permission slip.” Id. at 957. Judge Clay again seeks to distinguish Brown on this point, arguing that the Act “says nothing about what subject matter Children Users may access.” Clay Op. at 22. But the Act carves out operators whose users interact in the comment sections of “established and widely recognized media outlet[s].” Ohio Rev. Code § 1349.09(O)(2). As Judge Clay points out elsewhere in his opinion, the Act “favors news and current events subject matter.” Clay Op. at 20.

Moreover, the Act is ineffective—and therefore unnecessary for addressing the covered sites’ purported harms—due to the same verification problem alluded to in Brown. The Court in Brown noted that “one parent (or even an aunt or uncle)” could conceivably purchase a violent video game for a child, but the statute at issue did not have “any requirement[] as to how this parental or avuncular relationship is to be verified; apparently the child’s or putative parent’s, aunt’s, or uncle’s say-so suffices.” 564 U.S. at 802. Here, the Act allows covered operators to No. 25-3371 NetChoice, LLC v. Yost Page 70

register parental consent by one of five ways, including by signing a form, using a credit card, or calling “a toll-free telephone number . . . staffed by trained personnel.” Ohio Rev. Code § 1349.09(B)(1). But Yost cannot explain how these methods prove a parental relationship. For instance, how would an operator know that a person who called the toll-free telephone number is actually the prospective minor user’s parent? Apparently, the caller’s “say-so” is sufficient. Brown, 564 U.S. at 802. The Act’s approach to protecting children fails strict scrutiny.

2.

Yost also argues that the Act advances parents’ ability to make decisions involving their children’s care and upbringing. But the Act “is seriously overinclusive because it abridges the First Amendment rights of young people whose parents . . . think [social media is] a harmless pastime.” Id. at 805. Although “some of the legislation’s effect may indeed be in support of what some parents of the restricted children actually want, its entire effect is only in support of what the State thinks parents ought to want.” Id. at 804. “Not all of the children who are forbidden [from using covered websites] on their own have parents who care whether they [use covered websites].” Id.

In finding that the Act survives strict scrutiny, Judge Clay argues that “Ohio seeks to involve parents because their involvement itself will inherently mitigate some of the harms of unsupervised social media use.” Clay Op. at 23. But it is hard to see how one-time parental consent mitigates such harms unless those parents categorically prohibit their children from using social media on the front end. After all, the Act does not require parents to supervise their children as they use the covered websites. Nor does it create mechanisms for parents to check in with their children about social media habits or to re-consent periodically. The Act only requires that parents are engaged in the beginning and in a binary way—they can either allow their children to use covered websites or not. This approach does not narrowly or effectively advance the goal of involving parents in children’s social media use.

3.

Finally, the Act is not narrowly tailored to promote the state’s purported interest in regulating contracts with minors. For example, the law is under-inclusive because it carves out No. 25-3371 NetChoice, LLC v. Yost Page 71

the websites of “established and widely recognized media outlet[s].” Ohio Rev. Code § 1349.09(O)(2). Yost does not explain why a minor user’s contract with, for example, The New York Times is permissible but a similar contract with Facebook, or a media outlet that is not “widely recognized,” would not be. Furthermore, the Act is not narrowly tailored because, again, it uses too blunt of an instrument. The state could have prohibited operators from contracting with minors using particularized objectionable terms. Instead, the Act simply prevents minors from contracting with NetChoice’s members, cutting off their access to covered operators altogether.

IV.

For these reasons, I would find that the Act fails strict scrutiny. And because I would uphold the district court’s grant of summary judgment to NetChoice on this basis, I do not reach NetChoice’s vagueness arguments.

The district court also granted a permanent injunction against enforcement of the Act. Although I agree with the district court’s analysis of the permanent injunction factors, see Yost, 778 F. Supp. 3d at 958-59, I would narrow the injunctive relief to NetChoice, its members, and its members’ users. “The ‘scope of injunctive relief’ that a court may grant to a plaintiff depends on the scope of the constitutional ‘violation’ that the plaintiff has ‘established’ on the merits.” Doe, 165 F.4th at 530 (quoting Califano v. Yamasaki, 442 U.S. 682, 702–03 (1979)). Although courts may only grant relief to the violation “that produced the injury in fact that the plaintiff has established,” Lewis v. Casey, 518 U.S. 343, 357 (1996), in the context of a facial challenge, “courts may grant injunctions that prohibit the government from enforcing the law against the plaintiff in any circumstance—not just as applied to specific conduct” where a plaintiff has established “law’s unconstitutionality in the required number of applications.” Doe, 165 F.4th at 530. For me, NetChoice has made that showing here. See id.