Has Meta Met Its Match?

The new battle to regulate social media

Social media companies can’t catch a break. Throughout the past decade, representatives from Facebook, Instagram, TikTok, YouTube, and X have been on constant rotation through courtrooms and congressional hearings to defend their platforms. But even in the face of data scraping, investor fraud, monopolistic practices, national security threats, and rampant misinformation coming to light, children are still social media’s most ubiquitous victims [1] [2] [3] [4] [5]. Backed by new research, thirty-three states filed a joint lawsuit in October 2023 accusing Meta (the owner of Facebook, WhatsApp, and Instagram, among other platforms), of using unsafe consumer practices to target minors [6]. This lawsuit’s strategy—targeting addictive algorithms rather than the content they host—sidesteps outdated policy in its fight for a child-friendly digital world.
The real-world harms of online content are easy to identify. Doctor’s visits addressing teenagers with eating disorders more than doubled between 2018 and 2022, an increase that many experts blame on social media [7]. When Senator Richard Blumenthal from Connecticut created a fake Instagram persona impersonating a thirteen-year-old girl and dabbled in dieting content, his feed began promoting extreme weight loss within the week, recommending accounts like “eternally starved” and “I have to be thin” [8]. Kids are easily coerced by creators and trends, yielding deadly results. The notorious “Cinnamon Challenge,” wherein participants ingested spoonfuls of ground cinnamon, caused lung damage for dozens of children between 2007 and 2015 [9]. An even deadlier “Blackout Challenge,” which featured videos of participants choking themselves with belts, shoelaces, and other items to obtain an adrenaline rush, claimed the lives of at least twenty American children as of 2022 [10]. That same year, a British coroner notably ruled that Instagram and Pinterest had contributed “in a more than minimal way” to the suicide of fourteen-year-old Molly Russell, whose online accounts were full of self-harm and suicidal ideation, much of which she had copied verbatim into her journal [11]. As reporter Adam Satariano notes, the Russell family’s case was “perhaps the first time anywhere that internet companies [had] been legally blamed for a suicide.” In light of—sometimes deadly—consequences, concerned citizens are clamoring for heightened protections. 

While countries like Britain can hold platforms directly liable for third-party content, Section 230 of the Communications Decency Act prevents the United States from following suit [12]. When American attorneys pounced on the Russell ruling as a potential precedent for similar tragedies, platforms found a shield in this provision, which was implemented in 1996 to protect a budding Web 1.0 from legal challenges. Section 230(c)(1) states that:

“No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider.” 

The internet has grown exponentially since 1996, but Section 230’s immunity still extends far beyond the typical corporate standard of care. A grocery store chain would be tried for negligence if executives knew that an employee was poisoning the food. On digital platforms, however, individuals are solely liable for everything from cyberbullying to solicitation of minors—even if the platforms in question are aware of illegal activity. In an effort to fill this gap, law professors Danielle Keats Citron and Benjamin Wittes suggest qualifying Section 230’s immunity:

“No provider or user of an interactive computer service that takes reasonable steps to address known unlawful uses of its services that create serious harm to others shall be treated as the publisher or speaker of any information provided by another information content provider in any action arising out of the publication of content provided by that information content provider” [13].

 These addendums would require, at minimum, specific attempts by platforms to mitigate suspect material, like taking down explicit harassment or blocking searches for dangerous challenges. The latter could have saved poison control centers hundreds of frantic calls when the Cinnamon Challenge was trending [14]. 

But despite hopes for change, the Supreme Court has been reluctant to address Section 230 directly since Zeran v. American Online, Inc. cemented its supremacy in 1998. Plaintiff Zeran had filed the suit after an anonymous American Online Inc. (AOL) user ran a targeted harassment campaign against him, but the Supreme Court found AOL not responsible and Section 230 held firm [14]. Many believed that the Court would reconsider its stance decades later in Gonzalez v. Google and Twitter, Inc. v. Taamneh, considering the rapid evolution of internet platforms since 1998 [16] [17]. In May 2023, however, the Court’s decisions unanimously sidestepped considerations of publisher accountability, leaving Section 230 to continue operating as is. As of today, the Supreme Court doesn’t seem amenable to changing the technical wording of the advocates hoping to challenge Meta’s immunity through changing the word of Section 230, therefore, are fighting an uphill battle.

Beyond, Citron and Wittes’ amendments don’t do anything to address the scourge of destructive—but perfectly legal content—inundating kids’ phone screens. According to the states’ 2023 lawsuit, social media’s most universal harm lies not in its content, but in the platforms themselves. Apple’s release of the iPhone coincided with the steepest spikes in teen loneliness, sleep deprivation, and time spent without friends that psychology professor Jean M. Twenge had ever seen [18]. According to her book, “Teens who spend three hours a day or more on electronic devices are 35 percent more likely to have a risk factor for suicide”—regardless of the online material they’re viewing. Other studies suggest that suicidal ideation rises after just two hours per day [19]. She reports that “there’s not a single exception. All screen activities are linked to less happiness, and all non-screen activities are linked to more happiness.” Algorithms pull us not only from the real world, but also from mental health and happiness. For Facebook, at least, this association is causative; using Facebook generates sadness, but sadness does not lead to using Facebook [20][21]. Against this context, daily screen time averages—over five and a half hours for tweens aged 8-12 and over eight and a half hours for teens aged 13-18—are terrifying [22].

Meta, however, isn’t concerned. Despite a purported ban on kids under thirteen, the company collects profitable data from millions of confirmed underage users, banning a mere fraction of their accounts [23]. Furthermore, according to the lawsuit, Meta “designed psychologically manipulative product features,” including infinite scrolling and persistent notifications, “to induce young users’ compulsive and extended use.” In her 2021 testimony before Congress, whistleblower and former Facebook product manager Frances Haugen asserted that Facebook knew that as “young women begin to consume…eating-disorder content… [they] use the app more” [24]. As Senator Blumenthal’s experiment shows, algorithms inundate young women with increasingly extreme content, fueling obsessive usage. Addictive features have been so successful that the Attorneys General for both Nebraska and Colorado compared Big Tech to Big Tobacco, claiming that “Meta chose to maximize its profits at the expense of public health, specifically harming the health of the youngest among us” [25][26].

Regulating addictive algorithms might only require reinterpreting Section 230 as opposed to altering it. The provision’s wide reach is a result of Force v. Facebook, a 2019 federal appellate case responsible for the precedent that “Section 230(c)(1) should be construed broadly” [27]. Judge Robert Katzman disagreed. After being outvoted 2-1 by his fellows, he wrote a thirty-five-page long dissent for Force arguing that recommendation algorithms should not qualify for Section 230 protections: “It strains the English language to say that in targeting and recommending these writings to users…Facebook is acting as "the publisher of ... information.” In 2020, Justice Clarence Thomas affirmed this interpretation, asserting that “courts have…departed from the most natural reading of the text by giving Internet companies immunity for their own content” [28]. As Thomas sees it, algorithms select posts for publication, “add commentary, and then feature the final product prominently over other submissions… To say that… does not ‘creat[e] or develo[p]’ the final product, even in part, is dubious.” Recommendation algorithms push social media platforms outside the boundaries of “publisher or speaker” role that Section 230 references. In 2024, social media platforms are not publishers, but choosy and highly attuned amplifiers. Under this framework, Instagram’s role in Molly Russell’s suicide lies not in exposing her to suicidal content, but in repeatedly exposing her to suicidal content. If the data prove that young kids are too addicted to social media to stop scrolling, then the magnification of harmful material is an even bigger problem than the material itself. 

Fortunately, adapting algorithms for minors is not a ground-zero endeavor. In 2018, the United Kingdom pioneered a new model called the “children’s code,” which requires platforms to conform to an “age-appropriate” experience for kids or face fines [29]. Features range from placing curfews on notifications to turning off autoplay on videos. While these adjustments may seem insignificant—after all, the millions of minors addicted to Instagram don’t need night notifications to attract them to the app—they clear a path for more substantive changes down the line. In an ideal world, habit-forming elements like infinite scroll would be limited, children younger than twelve would be barred from creating accounts, and social media companies would need parental permission to collect and store their kids’ data [30]. Implementing a children’s code in the United States is the first step towards that reality.

Regardless of its outcome, this lawsuit will no doubt generate awareness about dangerously addictive algorithms. Already, Colorado and Tennessee are leading a 46-state coalition to dispute TikTok’s lack of consumer protections for minors in a similar challenge [31]. Judge Katzman articulated that Meta’s algorithm have done their job “too well, nudging susceptible souls ever further down dark paths” [32]. Hopefully, algorithmic regulation will be able to shed some light.