Social media is a drug. We have a basic biological imperative to connect with other people … there’s no doubt that a vehicle like social media, which optimizes this connection between people, is going to have the potential for addiction.
– Dr. Anna Lembke 2020Footnote 1
I. Introduction
In the span of just two decades, social media has transformed from a novel communication platform into an omnipresent force in modern life, fundamentally altering how we connect, communicate, and consume information. For adolescents born into this digital landscape, social media is not merely a tool but an integral component of identity formation, social development, and daily existence. Most teens ages thirteen to seventeen use social media daily, and over forty percent of Instagram’s user base is under the age of twenty-two.Footnote 2 Yet as platforms have evolved to capture and maintain user attention with increasing sophistication, alarming patterns have emerged that mirror addictive behaviors typically associated with substances and gambling. Medical research reveals that social media use can cause neurological changes with striking similarities to those seen in substance use disorder and behavioral addictions.Footnote 3 Some levels of social media use would be more accurately described as social media addiction, and social media use is associated with depression, anxiety, attention and sleep problems, body image issues, and low self-esteem.Footnote 4 The neurological, psychological, and social implications of social media use pose particular concerns for adolescent (young adult) users, who are especially vulnerable to the pernicious effects of social media.Footnote 5 The tech industry has capitalized on this by subjecting young users to addictive features designed to capture and retain their focus.Footnote 6 By way of data collection and personalized advertising, they have turned user attention into billions of dollars of revenue.Footnote 7 The young age at which users begin using social media, sometimes in ways that are problematic or addictive, makes clear that this is not simply a way of life in the digital age — it is a public health crisis that demands legal attention.
In recent years, this legal attention has been mounting. Countries around the world, states across the United States, and numerous individuals and elected officials are taking action to address the harms of social media to minors.Footnote 8 Social media bans have been passed and implemented, with Australia’s social media ban for those under sixteen years old becoming effective in December 2025.Footnote 9 Multiple states in the United States have passed legislation to prohibit addictive design features for minors online, and a coalition of state attorneys general have brought lawsuits against Meta for negligent design and harm to children.Footnote 10 Some of these children have grown up on social media, using platforms like YouTube as early as six years old, and have secured multi-million-dollar jury awards in negligence suits.Footnote 11 The increasing recognition of the negative effects of social media use on children and adolescents and the growing political will for regulation have started to change the online landscape, though federal legislation in the United States remains elusive.Footnote 12 Still, states and individuals have demonstrated that they can apply serious pressure to some of the largest companies in the world through legislation and litigation, providing multiple potential avenues to holding social media companies accountable in an age where social media is ubiquitous, especially for younger generations. The complexity of regulating social media companies poses substantial hurdles and inevitable litigation, but a path forward is both feasible and exigent. This Note examines the intersection of social media prevalence and negative mental health outcomes, the psychology behind addiction and addictive design, and legal and technical frameworks to address a critical question: How can the law, through legislative, judicial, or regulatory means, protect adolescent users from the harmful effects of social media addiction?
First, this Note will define the problem of social media addiction, the associated mental health risks, the prevalence of social media use among adolescents, and the business model of large social media platforms. Second, I will examine the neurological pathways of addiction and the ways that social media platforms exploit these pathways through addictive design and dark patterns. Third, I will analyze the current legal frameworks that constrain and enable possible regulatory approaches and discuss novel legal solutions. Finally, I will summarize proposed company-led and third-party technical solutions to limit the addictive qualities of social media for adolescent users.
This Note will argue that the harms of social media to adolescent users should be addressed by directly regulating the addictive design choices of social media platforms. Given potential challenges under the First Amendment and Section 230 of the Communications Decency Act, legislation aimed at preventing harm and addiction should focus on limiting or prohibiting these features, including algorithmic recommendations, notifications, infinite scroll, and autoplay, rather than regulating the content hosted on these platforms. Until such legislation is effective, redress for past and ongoing harm is best achieved through negligence suits against social media companies that are knowingly implementing addictive features, marketing their platforms to children, and misleading the public about the safety of their platforms. A multi-pronged approach through the courts and the legislature presents the best opportunity to hold social media companies accountable for harming children and preventing future generations from facing negative mental health outcomes and addiction.
II. Social Media Addiction
A. Definition and Scope
Internet addiction has been researched almost as far back as 1993, when the release of the World Wide Web made the internet widely accessible to the public.Footnote 13 The first findings on internet addiction were presented at the American Psychological Association (APA) just three years later, relying on criteria adapted from those used to diagnosis gambling addiction.Footnote 14 Since the first study in 1996, numerous studies have been conducted on internet addiction.Footnote 15 Within this concept of internet addiction lie more traditionally recognized addiction disorders, such as gambling addiction and shopping addiction, as adapted to the internet age.Footnote 16 Over the past two decades, with the rise of social media and technological advances which enable individuals to access the internet with fewer barriers than ever before, the landscape of internet addiction has shifted dramatically.
As the use of social media has exploded over the past decade, concerns about internet addiction have sharpened to focus especially on addiction to social media. danah boyd and Nicole Ellison, early researchers in social media, established a definition for social network sites in their foundational 2007 paper: “web-based services that allow individuals to (1) construct a public or semi-public profile within a bounded system, (2) articulate a list of other users with whom they share a connection, and (3) view and traverse their list of connections and those made by others within the system.”Footnote 17 Social media is an umbrella term that may encompass platforms beyond social network sites, however for the purposes of this paper I will use these terms interchangeably.
The American Society of Addiction Medicine defines addiction as “a treatable, chronic medical disease involving complex interactions among brain circuits, genetics, the environment, and an individual’s life experiences. People with addiction use substances or engage in behaviors that become compulsive and often continue despite harmful consequences.”Footnote 18 While there is no prevailing medical definition specifically for social media addiction, a group of researchers from the University of Hong Kong, the University of Oslo, and the National Institute on Alcohol Abuse and Alcoholism put forth a useful definition: “Social media addiction is a type of behavioral addiction that is broadly defined as compulsive engagement in social media platforms that significantly disrupts the users’ functioning in important life domains, such as interpersonal relations, work or study performance, and physical health.”Footnote 19 The researchers listed six types of problematic behavior to conceptualize social media addiction — salience, or dominance of social media in one’s daily life; tolerance, resulting in the tendency to increase the time spent on social media to achieve the same level of pleasure; mood modification, or use of social media to avoid unpleasant emotions from real-life events; relapse, or failure to curb social media use; withdrawal symptoms from not using social media; and conflict, or adverse impact in one’s professional or academic life.Footnote 20 By this definition, social media addiction, like any other form of addiction, is a harm in itself.
With the prevalence of social media in the United States, especially among minor users, many — including the U.S. Surgeon General — have called for increased research into the area.Footnote 21 In 2013, the APA recognized Internet Gaming Disorder in the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) as “a condition warranting more clinical research and experience before it might be considered for inclusion … as a formal disorder.”Footnote 22 In a 2022 study of 1,047 participants from the United Kingdom and the United States, researchers found that the risk of social media addiction was two to four times higher for U.S. adults aged eighteen to thirty-four years old compared to U.S. adults aged fifty to sixty-five.Footnote 23 The difference was even starker in the U.K. group, with young adults at a four times higher risk compared to older adults.Footnote 24 The heightened vulnerability of young adults, not to mention adolescent users, underscores the urgency of legislation and regulation in the social media sphere.
B. Mental Health Harms Associated with Social Media Use
Social media addiction has been the subject of numerous research papers, ranging from behavioral surveys to neurological imaging studies, however a causal link between social media use and a wide range of adverse health outcomes has yet to be established. In 2023, the Surgeon General of the United States issued an advisory report titled “Social Media and Youth Mental Health.”Footnote 25 The report outlined the widespread use of social media by children and adolescents and the correlations with adverse mental health outcomes.Footnote 26 In particular, the report pointed to sleep problems, attention problems, depression, anxiety, low self-esteem, and poor body image.Footnote 27 These issues have been well-documented in the context of social media use.Footnote 28
A recent study conducted with university-age students in Germany imposed social media abstinence for a fourteen day period and assessed measures of problematic smartphone use (PSU), depression and anxiety, and body image dissatisfaction.Footnote 29 The researchers found that social media abstinence was associated with less overall screentime and decreased body image dissatisfaction.Footnote 30 Another study limited undergraduate students’ social media use to ten minutes for each of three platforms (Facebook, Instagram, and Snapchat) per day, and found that these students showed a significant reduction in loneliness and depression at the end of three weeks as compared to a control group that placed no limitation on use.Footnote 31 Further research has demonstrated a positive link between social media addiction and eating disorders,Footnote 32 and analysis of content on platforms including Facebook, TikTok, Instagram, and X (formerly Twitter) revealed consistent themes of promoting “extreme thinness, providing tips on weight loss, and fostering a positive sentiment towards low body weight.”Footnote 33 There is also a growing body of research examining the coinciding increase of youth suicide rates and social media use, finding that social media has the potential to heighten or mitigate self-harm risk, but the amount of time spent on social media may have a direct correlation with self-harm behavior and suicidal ideation.Footnote 34
Furthermore, many of the adverse outcomes associated with social media addiction mirror those associated with other types of addiction, including mental health issues as well as neurological changes.Footnote 35 Research shows heightened social media use is correlated with smaller gray matter volumes in different regions of the brain.Footnote 36 Decreased gray matter is associated with lower impulse control and is similar to the morphological changes in the brain seen with substance or gambling addiction.Footnote 37 This may be in part because social media companies and casinos target the same underlying psychological processes to maximize user engagement, and indeed social media appears to have taken inspiration from different gambling games, such as slot machines.Footnote 38
The evidence found in behavioral studies, neurological research, and public health assessments demonstrates that social media addiction is a significant concern across disciplines. The documented impacts of social media have striking overlap with other forms of addiction, though further research is needed to establish a causal relationship between these adverse effects and social media.
C. Prevalence of Social Media Use Among Adolescents
The growing body of research on social media use demonstrates that adolescents and young adults make up a significant portion of the user base for leading social media platforms. As of 2021, over forty percent of Instagram users were twenty-two years old and younger.Footnote 39 Eighty percent of adults aged eighteen to twenty-nine have used Instagram, along with ninety-five percent for YouTube, sixty-eight percent for Facebook, and sixty-three percent for TikTok.Footnote 40 The percentage of users in this age group is higher compared to older age groups for nearly all eleven social media platforms surveyed, with the only exceptions being Facebook, WhatsApp, and Truth Social.Footnote 41 In addition to the high rates of youth and young adult use, adolescent users may be particularly vulnerable to the negative effects of social media use.
A 2024 study from Pew Research Center revealed that most teens ages thirteen to seventeen use social media daily.Footnote 42 The most popular platforms among this age group are YouTube (ninety percent of teens reported any use), TikTok (sixty-three percent of teens), and Instagram (sixty-one percent of teens).Footnote 43 A majority of teens who reported any use of these platforms also reported using them daily, with most of this subset using YouTube, TikTok, and Instagram “several times a day.”Footnote 44 In 2025, the statistics stayed relatively stable, and across the top five platforms (YouTube, TikTok, Instagram, Snapchat, and Facebook), thirty-six percent of teens reported using at least one site “almost constantly.”Footnote 45 The 2025 report also included statistics on the use of AI chatbots for the first time, finding that a majority of teens use chatbots, with twenty-eight percent of teens reporting daily usage.Footnote 46 Use of AI chatbots is another area of heightened concern for both adolescents and adults, as news reporting has found that AI chatbots can lead to psychosis, suicidality, and violence.Footnote 47 With social media platforms beginning to incorporate AI chatbots into their applications, this may become a greater focal point in the coming years.Footnote 48
Common Sense Media, a nonprofit organization focused on technology, education, and digital equity, conducted studies in 2019 and 2021 on social media use for tweens (ages eight to thirteen) and teens (ages thirteen to eighteen).Footnote 49 The 2021 study incorporated the question of whether total screen media use had changed since the onset of the COVID-19 pandemic, and found that since the 2019 report, initiation into social media use is trending younger.Footnote 50 In fact, the percentage of tweens using social media every day and the average amount of time spent per day have both increased since 2019.Footnote 51 The study also reported that “while the time devoted to social media among teens may be increasing, enjoyment is not.”Footnote 52
Meta, the parent company of Facebook and Instagram, has also conducted internal research on the impacts of Instagram on its user base. While Meta refused to make their research available to the public, academic researchers, or lawmakers, their findings became publicized after a former Facebook employee, Frances Haugen, leaked the files to The Wall Street Journal in 2021.Footnote 53 Meta’s research makes clear that the company, including CEO Mark Zuckerberg, is aware of Instagram’s negative effects on teen users, and deliberately obfuscated such knowledge from the public.Footnote 54 Internal documents and research from Meta have become important pieces of evidence in lawsuits against the company, including two recent cases that resulted in multi-million dollar jury verdicts against Meta.Footnote 55
Among the discoveries in the leaked documents was Project Daisy, a years-long experiment to determine if hiding “like” counts on posts improved the negative feelings that users associated with “like” counts.Footnote 56 Facebook presented Project Daisy to Zuckerberg in 2020 and found no observable movements in “overall well-being measures” when “like” counts were hidden, yet Meta rolled out this change to Facebook and Instagram in May 2021 “after senior executives argued to Mr. Zuckerberg that it could make them look good by appearing to address the issue [of anxiety and negative feelings].”Footnote 57
Meta’s knowledge of mental health issues, either created or worsened by their social media platforms, and its attempt to conceal this information from the public, has generated comparisons to Big Tobacco’s efforts to market harmful and addictive products to children while spreading denialist propaganda about the dangers of smoking.Footnote 58 Senator Richard Blumenthal (D-CT) criticized Facebook’s evasion of congressional questions about the impact of Facebook and Instagram on child and adolescent wellbeing: “Facebook seems to be taking a page from the textbook of Big Tobacco—targeting teens with potentially dangerous products while masking the science in public.”Footnote 59 The pathways of addiction and the neurological similarities between social media addiction and other substance and behavioral use disorders is further explored in Part II.
The existing data paints a clear picture of adolescents as both the primary users and most vulnerable consumers of social media. With platforms like YouTube, TikTok, and Instagram dominating teen digital life, and usage patterns trending toward younger initiation and more frequent engagement, the stakes of this technological experiment on youth development continue to rise. There is a concerning parallel between social media companies and Big Tobacco in terms of targeting young users and obscuring evidence of harm, which raises serious legal questions of regulatory oversight and potential liability for these companies. As social media becomes more embedded in adolescents’ everyday life, the urgency of addressing its harms only increases.
D. The Business Model of Social Media
The most popular social media platforms among adolescents — YouTube, Instagram, and TikTok — rely primarily or entirely on advertising revenue to generate profits. Personalized advertisements are more cost-effective, as they can increase revenue while decreasing marketing costs.Footnote 60 To push personalized advertisements, social media platforms collect vast amounts of user data to build algorithms that can easily and quickly identify which product or brand to promote to any given user.Footnote 61
This type of business model is part of the larger trend towards information accumulation in the digital age, which Shoshana Zuboff calls surveillance capitalism: “This new form of information capitalism aims to predict and modify human behavior as a means to produce revenue and market control.”Footnote 62 The desired human behavior in this case is online user engagement. The bigger picture, however, reveals that it’s not just users’ attention that social media companies profit off, rather it is “the gradual, slight, imperceptible change in your own behavior and perception that is the product.”Footnote 63 The ability to manipulate or direct user behavior becomes an immensely profitable tool that social media companies can sell to advertisers over and over; as Zuboff has articulated, “we now have markets that trade in human futures at scale, and those markets have produced the trillions of dollars that have made the Internet companies the richest companies in the history of humanity.”Footnote 64
In order to push as many ads as possible and maximize revenue, social media platforms need to capture and maintain user engagement. They do so through attentional scaffolds, or processes that “systematically influence users’ focus, management, and direction of attention, either benefitting or harming their ability to achieve their goals.”Footnote 65 Social media platforms function as hostile attentional scaffolds by taking advantage of users’ emotions and employing addictive design features, often called dark patterns, benefitting platforms at the expense of their users.Footnote 66 On a societal scale, this ability to manipulate and guide user behavior has widespread implications. Beyond its impact on individuals’ mental health, social media has the potential to sow political division, encourage authoritarianism, and influence election outcomes.Footnote 67
III. Pathways of Addiction & Social Media Design
A. The Neurological Model of Addiction
Social media, similar to addictive substances, significantly impacts brain chemistry and morphology. Understanding the biological mechanisms behind social media addiction requires examining its effects on the brain at both structural and chemical levels. Recent studies have revealed fascinating similarities between the neural changes associated with social media use and those observed in substance use disorder, suggesting common underlying pathways of dependence which may grant insight into the ways that social media platforms create and reinforce addictive patterns of behavior.Footnote 68
Neuroimaging studies on the use of social media have revealed links between heightened social media use and abnormalities in brain morphology, particularly in volume of white matter and gray matter.Footnote 69 These changes have been measured in the prefrontal cortex, the amygdala, and the nucleus accumbens, regions which are involved in reward, executive attention, and decision-making.Footnote 70 Disruptions in the prefrontal cortex can lead to impaired response inhibition, making destructive behaviors such as drug seeking a motivational drive even at the expense of non-drug behaviors.Footnote 71 Findings of abnormal gray matter volume in the brain’s prefrontal regions are common in individuals suffering from use disorders involving nicotine, heroin, alcohol, opiates, methamphetamines, and cocaine.Footnote 72 The similarities in structural changes to the brain indicate that social media addiction may be facilitated by the same neural pathways as substance use disorder, an idea that is further supported by chemical similarities.
One of the neurochemical pathways involved in addiction is dopamine activity in reward signaling, which innervates the nucleus accumbens, a structure in the brain that is involved in pleasure, reward, pain, and addiction.Footnote 73 Dopamine is a neurotransmitter responsible for feelings of pleasure, and is involved in reinforcement, sleep, learning, attention, mood, and pain processing.Footnote 74 Reduction of dopamine receptors has been observed in cocaine, cannabis, and alcohol use disorders along with morbid overeating behavior.Footnote 75 This reduction may contribute to an increased need for high levels of excitement and stimulation, which may be achieved through internet and social media use.Footnote 76 Research in monkeys has shown that where rewards were maximally uncertain — that is, where a reward followed a signal only fifty percent of the time — dopamine was released at the highest magnitude.Footnote 77 Psychiatrist Anna Lembke posits that social media use causes a large dopamine release in users, but “[u]pon signing off, the brain is plunged into a dopamine-deficit state as it attempts to adapt to the unnaturally high levels of dopamine social media just released.”Footnote 78 Social media use, like many types of drug use, brings a dopamine surge that over time blunts dopamine receptors, creating a cycle of heightened and depressed dopamine that results in users going back to social media to get their dopamine hit.Footnote 79
Oxytocin, often referred to as the “love hormone,” may also play a role in encouraging social media use from a chemical perspective, though this has been less well documented. Oxytocin promotes trust, relaxation, and fosters social bonds.Footnote 80 Psychologist Paul Zak has found that ten minutes of social media use results in increased levels of oxytocin in the bloodstream by up to thirteen percent.Footnote 81 Oxytocin may come into play especially when users are interacting with their loved ones through social media, as suggested by Zak’s anecdote about one young study participant whose oxytocin levels increased by 150% after posting to his girlfriend’s Facebook page for ten minutes.Footnote 82
There is a compelling case for examining social media addiction through a neurological lens. The observed structural changes in gray matter volume, particularly in those regions associated with reward and executive function, mirror abnormalities in gray matter volume commonly seen in substance use disorders.Footnote 83 The chemical side of addiction, notably involvement of key neurotransmitters like dopamine and oxytocin, suggests that social media platforms may be tapping into the same reward and decision-making pathways that evolved to regulate social bonding and reward-seeking behavior. As is the approach in substance use disorder and other behavioral addiction, social media addiction should be addressed by way of legal regulation and medical intervention, and social media companies that intentionally implement addictive pathways should face oversight and liability.
B. Social Media Design Exploits Neurological Pathways
The neurological impacts of social media use are not merely incidental. Instead, they are the product of intentional design choices that exploit natural reward circuits in the brain. Sean Parker, the first President of Facebook, described Facebook’s design features as “a social-validation feedback loop … exploiting a vulnerability in human psychology.”Footnote 84 Facebook and other social media companies have developed sophisticated methods to trigger dopamine release and activate reward pathways by drawing from the same psychological principles used in gambling and online gaming.Footnote 85 By examining platforms’ specific design choices and their relationship to neurological pathways, we can better understand how social media companies deliberately exploit users’ psychology to maximize user engagement.
Many platforms employ something called the “Hook Model”, a four-phase feedback loop that exploits vulnerabilities in human psychology to create habitual users.Footnote 86 The model consists of “(1) a trigger that leads to (2) an action, which produces (3) a reward and ultimately creates (4) an investment.”Footnote 87 Others have described this feedback loop as a “compulsion loop,” that utilizes variable-rate reinforcement to manipulate dopamine pathways and increase the desired behavior.Footnote 88 From a psychology standpoint, rewards are any type of positive reinforcement that encourages specific behaviors.Footnote 89 Variable rewards are favored because research has shown that rewards are most effective at creating dopamine rushes when they are uncertain or unexpected.Footnote 90
The intermittent rewards come in the form of content and feedback.Footnote 91 Content serves to keep users engaged while they’re online, which can be more effectively achieved or prolonged through algorithmic content recommendations.Footnote 92 Feedback includes likes, comments, shares, and other engagement metrics. Research has found that users’ social media activity increases after they post, possibly in anticipation of rewards.Footnote 93 Users are also vulnerable to social reciprocity — users will provide more feedback to others after receiving feedback themselves.Footnote 94
C. Addictive Design Features
In granting rewards to users, platforms employ a variety of design choices, primarily focused on getting users online and creating a seamless user experience to maximize the time spent online. Many platforms rely on operant conditioning, a psychological process where behaviors are shaped through rewards and consequences. Each tap, scroll, and interaction teaches users to associate certain actions with specific rewards, whether it’s a notification, a like, or fresh content. This conditioning process is not incidental but rather central to how these applications function and maintain user engagement.
In conditioning users, social media companies designed certain features with slot machines in mind.Footnote 95 Slot machines, in turn, operate much like a Skinner box, named for B.F. Skinner’s foundational study in operant conditioning.Footnote 96 In the simplest configuration, a rat is placed inside a box with a lever. When the rat pushes the lever, food comes out, positively reinforcing the action of pushing the lever. Then, the reward of food changes from constant to intermittent and variable. When the rat pushes the lever, it may receive no food, or it may receive a variable quantity of food. The rat does not know when it will receive a reward, and thus will push the lever over and over, even when it receives nothing: “the rat becomes obsessed—addicted, if you will. This, then, is the psychological principle that slot machines operate on, and how it operates on you.”Footnote 97
These “levers” are present in nearly every digital app, even those that don’t fall within the definition of social network sites. Tristan Harris, former Design Ethicist at Google and co-founder of the Center for Humane Technology, explained that users are playing a slot machine every time “we pull our phone out of our pocket … we pull to refresh our email … we swipe down our finger to scroll the Instagram feed … we swipe faces left/right on dating apps like Tinder … we tap the # of red notifications.”Footnote 98 Digital apps, by their very nature, operate within a sophisticated reward system that strengthens behavioral patterns through both immediate and variable reinforcement.
One important facet of this reward system is notifications, which encourage users to look at their phones and check social media. To exploit the dopamine surge that uncertainty produces, notifications are sent at irregular intervals and frequency.Footnote 99 These notifications can induce fear of missing out (FoMO), or a “pervasive apprehension that others might be having rewarding experiences from which one is absent.”Footnote 100 Researchers postulate that FoMO includes two processes: (1) perception of missing out; and (2) compulsive behavior to maintain social connections.Footnote 101 Furthermore, FoMO is associated with a range of negative health outcomes, including lack of sleep, anxiety, and lack of emotional control.Footnote 102 FoMO is also triggered by time-sensitive features, such as story posts that disappear after twenty-four hours.
Some apps built features almost entirely around this concept, such as Snapchat, which requires daily use to maintain “streaks,” and BeReal, an app marketed as an “organic, person-to-person form of social media.”Footnote 103 BeReal notifies users at a random time every day, starting a two-minute window for users to take a picture to be posted. BeReal informs users that “you can only see what your friends are doing if you post in time.”Footnote 104
Once users are on social media, a number of user interface and user experience (UI/UX) design features work to keep users engaged. One such design feature is infinite scroll, which is incorporated into virtually every modern social media platform.Footnote 105 Developed in 2006 by Aza Raskin and intended to reduce friction for users, infinite scroll loads content as the user scrolls such that they never have to wait for new content to load.Footnote 106 This enables the user to sit passively while being fed seemingly endless amounts of content. In Raskin’s words, “if you don’t give your brain time to catch up with your impulses, you just keep scrolling.Footnote 107 Infinite scroll, combined with algorithmic recommendations, creates a time sink where users get lost in their own virtual world. This hyper-focused yet isolated environment is incredibly similar to the theory of architectural design in casinos. Sociologist Gerda Reith described the intentional flattening of space and time in the context of gambling: “The surrounding space of the periphery is obliterated … [space] shrinks into a single point and loses its extensity.”Footnote 108 Author Natasha Dow Schüll extends this explanation, writing that the single point of focus is “the device’s screen—and it is less a point of spatial reference than it is a point of exit from space.”Footnote 109 Smartphones, and specifically social media apps, operate in this exact way.
These design features may be more aptly labeled as dark patterns, defined as “user interface design choices that benefit an online service by coercing, steering, or deceiving users into making decisions that, if fully informed and capable of selecting alternatives, they might not make.”Footnote 110 In the words of Tristan Harris, “our choices are not as free as we think they are.”Footnote 111 Through operant conditioning, social media platforms subtly manipulate user behavior to the benefit of the platforms and to the detriment of the consumer.
The deliberate architecture of social media platforms reveals a sophisticated understanding and exploitation of human psychology. From notifications that mirror slot machine mechanics to infinite scroll features that dissolve temporal awareness, these design choices are carefully engineered to maximize engagement through proven principles of behavioral conditioning. The implementation of variable reward schedules, FoMO triggers, and frictionless interfaces creates an environment explicitly designed to bypass users’ rational decision-making processes and exploit psychological vulnerabilities. Even newer platforms marketed as more authentic alternatives, such as BeReal, incorporate these dark patterns. The parallel between casino architecture and social media design is particularly telling, with both creating isolated environments that disconnect users from external reference points, transforming discrete moments into endless loops of engagement. The strategic manipulation of user behavior through design raises important questions about the ethical implications of platforms that deliberately exploit psychological vulnerabilities for commercial gain. With the growing body of psychology research against a backdrop of increasing attention on social media companies, some jurisdictions are charting a legal pathway to protect children from the harms of social media.
IV. Legal Frameworks & Obstacles In Social Media Regulation
In the past decade, there has been varied discussion of how to approach and address social media harms from a regulatory perspective. The overarching goal of regulation is to prevent social media platforms from profiting off an adolescent user base while subjecting users to harms of addiction, depression, anxiety, eating disorders, and more.Footnote 112 Stories of adverse health outcomes for adolescent social media users are becoming increasingly frequent and wide-ranging, with some ending in tragedy, and the existing scientific research seems to confirm the role of social media in self-harm behavior and even suicide.Footnote 113 Another significant goal is to hold tech companies accountable for the harms their products cause for adolescent users, and particularly those harms that the companies were aware of but neglected to mitigate.Footnote 114 To achieve these ends, legal approaches that focus on the addictive design of social media platforms offer the most promising regulatory pathway.
A. Obstacles to Regulation
To understand the feasibility of regulation, measures must be crafted with consideration of two major obstacles to regulating social media platforms in the United States: Section 230 of the Communications Decency Act and the First Amendment of the U.S. Constitution.
Section 230 of the Communications Decency Act contains safe harbor provisions exempting online platforms from liability for content created by third parties, stating 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.”Footnote 115 The statute contains significant exceptions to this exemption, including instances of copyright infringement.Footnote 116 In theory, Section 230 promotes free speech by exempting platforms from liability for content posted by users, allowing for a free and open internet.Footnote 117 In practice, however, Section 230 disincentivizes platforms to moderate harmful third-party content by shielding the platforms from liability for the impacts of such content.Footnote 118 While Section 230 is a significant consideration in the regulation of social media, there are ways around Section 230 immunity.Footnote 119 This Note will therefore focus primarily on what appears to be a greater obstacle to regulation — the First Amendment.
The First Amendment secures the right to free speech by prohibiting government abridgement and applies to both the federal government and states as incorporated by the Fourteenth Amendment.Footnote 120 In the context of social media regulation, the exercise of editorial functions and content moderation must be squared with the First Amendment rights of both users and platforms. Regulations are often separated into two categories: content-based, where the regulation treats expression differently based on its content, or content-neutral, where the regulation treats all expression the same but may instead target the time, manner, or place.Footnote 121
Laws aimed at regulating speech on the basis of its content are reviewed under strict scrutiny.Footnote 122 Strict scrutiny is the highest level of constitutional review under which a law is presumptively unconstitutional, thus creating a difficult hurdle to surmount.Footnote 123 The burden of proof rests on the government to show that (1) there is a compelling governmental interest that the law seeks to legislate; (2) the law was narrowly tailored in achieving its goal; and (3) the government exercised the least restrictive means in doing so.Footnote 124 Because strict scrutiny applies to laws that regulate speech, state-level legislative efforts to regulate online content have faced significant legal challenges and have often been enjoined by the courts.Footnote 125 The “least restrictive means” factor is where content-based laws typically fail to survive strict scrutiny.Footnote 126 For this reason, regulation of social media platforms should be crafted with particular care to the least restrictive means factor of strict scrutiny. Alternatively, regulation that targets addictive design rather than content itself may be able to get around the First Amendment strict scrutiny hurdle.Footnote 127
1. Moody v. NetChoice
In 2021, Florida and Texas enacted statutes that limited the content moderation ability of social media sites and internet platforms to combat perceived anti-conservative bias on these sites.Footnote 128 Among the statutory provisions were prohibitions on deplatforming political candidates and required disclosures about content moderation policies, to be enforced by fines and civil suit.Footnote 129 NetChoice, a national trade association of online businesses including industry giants like Google, Amazon, Meta, and X, among others, brought a First Amendment challenge.Footnote 130 The lower courts held that the law infringed on platforms’ content moderation choices, which constituted expressive speech and thus was protected under the First Amendment.Footnote 131 NetChoice secured preliminary injunctions preventing the laws from taking effect and the cases were jointly appealed up to the Supreme Court.Footnote 132 In July 2024, the Supreme Court agreed with the lower courts that editorial functions of social media platforms were no different from that of newspapers — content moderation constituted protected speech under the First Amendment.Footnote 133 The Supreme Court remanded both cases to the circuit courts for a complete analysis of the facial First Amendment challenges brought by NetChoice, reasoning that the government’s justification of “better balancing the marketplace of ideas” did not constitute a legitimate government interest, and holding that NetChoice failed to prove that the challenged laws were facially unconstitutional.Footnote 134 It remains to be seen whether the laws will stand. However, given the high standard of strict scrutiny for content-based regulations, it is difficult to imagine both laws will be upheld, and for this reason other states have not approached regulation of social media through content moderation.
B. Legislative Measures
Federal law on social media has lagged far behind advancements in technology, though some lawmakers have introduced legislation intended to regulate technological and design choices. In 2019, Senator Josh Hawley proposed the Social Media Addiction Reduction Technology Act (SMART Act), which would prohibit social media companies from using infinite scroll, auto refill, and autoplay functions; implement time restriction defaults; implement regular stopping points in scrolling; and notify users twice an hour of their daily time spent on the platform.Footnote 135 Senator Hawley pointed to the business model of these companies, stating that “Too much of the ‘innovation’ in this space is designed not to create better products, but to capture more attention by using psychological tricks that make it difficult to look away.”Footnote 136 The bill was referred to the Committee on Commerce, Science, and Transportation, but received little support in the Senate and has been stagnant since.Footnote 137
In April 2024, Senator Brian Schatz proposed the Kids Off Social Media Act (KOSMA), which would ban children under age thirteen from using any social media platform; prohibit recommendation algorithms for users under seventeen; and ban social media in schools.Footnote 138 This bill was also referred to the Committee on Commerce, Science, and Transportation, but failed to make it out of committee. However, Senator Schatz brought KOSMA forth again in January 2025, and this time the bill was reported favorably by the committee and placed on the Senate Legislative Calendar, though no further action has been taken since June 2025.Footnote 139
Given the difficulty of passing federal legislation and in light of the harms to adolescents in particular, many states have introduced novel legislation to protect youth within their state from the harms of social media. A myriad of approaches exist within these efforts, though many have been subject to intense litigation under First Amendment challenges by NetChoice, a national trade association of online businesses including industry giants like Google, Amazon, Meta, and X, among others.Footnote 140 In 2025, Arkansas passed a law creating a private right of action where a social media platform fails to use reasonable care in safeguarding against platform design, algorithms, or features that cause eating disorders, suicide, and platform addiction.Footnote 141 In December 2025, the Western District of Arkansas granted a preliminary injunction to NetChoice on the basis that the Arkansas law was not narrowly tailored enough to survive First Amendment strict scrutiny.Footnote 142
1. Warning Label Laws
Minnesota was the first state to require platforms to display warning labels about mental health with a bill passed in July 2025, following an opinion piece written by the U.S. Surgeon General advocating for warning labels, similar to those on cigarette packaging.Footnote 143 The law is set to go into effect in July 2026; however, NetChoice has already filed suit, claiming the warning labels are compelled government speech in violation of the First Amendment.Footnote 144 Since July 2025, three more states have passed warning label laws (Colorado, California, and New York).Footnote 145 While these laws have not yet been struck down, anticipated litigation may cause significant delay if not outright barriers to implementing warning label requirements.
Such litigation will likely allege that state laws requiring warning labels amount to compelled speech and violate the First Amendment. This is exactly the tack that NetChoice is taking in Minnesota, perhaps modeled after tobacco litigation, namely R.J. Reynolds Tobacco Company v. FDA. Footnote 146 Under the Family Smoking Prevention and Tobacco Control Act of 2009, Congress authorized the FDA to regulate the sale of tobacco products and required that cigarette packages include new warning statements and labels to comprise a certain percentage of the packaging.Footnote 147 The FDA then issued a Proposed Rule of nine images for potential use on the warning labels based on scientific research on the harms of tobacco and the effectiveness of graphic warnings compared to text-only warnings, justifying these new labels under the government’s “substantial interest in reducing the number of Americans, particularly children and adolescents, who use cigarettes and other tobacco products in order to prevent the life-threatening health consequences associated with tobacco use.”Footnote 148 R.J. Reynolds, along with four other tobacco companies, filed suit against the FDA, claiming that the warning labels amounted to compelled speech in violation of the First Amendment.Footnote 149
Under the First Amendment, different types of speech are subject to varying levels of review. Content-based speech regulations and compelled speech are subject to strict scrutiny.Footnote 150 Within content-based regulations and compelled speech, there are two exceptions for commercial speech.Footnote 151 First, for “purely factual and uncontroversial” disclosures that are not “unjustified or unduly burdensome,” they must be “reasonably related to the State’s interest in preventing deception of consumers.”Footnote 152 That is, where the government can prove the regulated actor’s commercial speech is misleading or deceptive, the regulation must only pass rational basis review. Second, the standard of review for commercial speech is intermediate scrutiny rather than strict scrutiny.Footnote 153 Under intermediate scrutiny, the government must affirmatively prove that (1) its asserted interest is substantial; (2) the restriction directly and materially advances that interest; and (3) the restriction is narrowly tailored.Footnote 154 To determine which standard to use for regulation of commercial speech, courts employ the Central Hudson test, which has four factors: (1) To warrant First Amendment protection, commercial speech must concern lawful activity and not be misleading; (2) the governmental interest must be substantial; (3) the restriction on speech must directly advance the governmental interest; and (4) limits on speech must be no more extensive than necessary to serve that interest.Footnote 155
In Reynolds, the Court reasoned that the FDA rule was not a remedial measure for “the years of deception [by tobacco companies] that preceded” the graphic warnings, nor was cigarette packaging itself misleading to consumers.Footnote 156 Furthermore, the graphic warnings were not purely factual and uncontroversial because they were intended to evoke an emotional response to deter cigarette use.Footnote 157 Therefore, the regulation did not meet the criteria for evaluation under Zauderer rational basis review, and was instead evaluated under Central Hudson intermediate scrutiny.Footnote 158 The Court found that the graphic warnings failed intermediate scrutiny as the government failed to bring substantial evidence the warnings would “directly advance” its governmental interest in reducing smoking rates.Footnote 159 While the success of warning labels in the social media context will require state laws to defend against similar First Amendment challenges, establishing warning labels as a remedial measure for platforms’ deceptive claims about safety or using simple, text-only warnings may help states in the inevitable litigation over warning label laws.Footnote 160
Given the likelihood of litigation and the difficulty of crafting a law to withstand such challenges, the most promising legislation is that which focuses on addictive design features rather than speech or content moderation. Two states that have so far been successful in regulating platform design are New York and California.
2. The New York Approach
One promising approach has emerged in New York. In June 2024, New York Governor Kathy Hochul signed the Stop Addictive Feeds Exploitation (SAFE) for Kids Act into law.Footnote 161 The SAFE for Kids Act targets the potential for addiction through algorithmic recommendations and notification systems and defines minors as those under age eighteen.Footnote 162 The New York law prohibits (1) use of minors’ personal information to target content through algorithmic recommendations, instead forcing social media companies to show minors chronological or non-personalized feeds, and (2) notifications between midnight and 6 AM ET.Footnote 163 The SAFE for Kids Act will not go into effect until 180 days following the publication of rules from the NY Attorney General, meaning that it could still be challenged in court.Footnote 164 The Attorney General published proposed rules in September 2025 with a public comment period of sixty days, after which the Attorney General has one year to finalize the rules and start the 180-day effective date clock.Footnote 165 However, given that the Act does not regulate content online, but merely prohibits use of algorithms and notifications for minors, it is less likely to face the First Amendment challenges received by platform-regulating legislation passed in Texas and Florida.
Platforms may argue that the design features targeted — algorithmic recommendations and notifications — are also protected under the company’s First Amendment speech rights as a form of expressive speech and content moderation. This means that under Moody v. NetChoice, regulation of design features would be subject to strict scrutiny. However, this argument hasn’t been tested, and in other cases, negligence claims around the harms of algorithmic recommendations haven’t been barred by the First Amendment.Footnote 166
3. The California Approach
Similar to New York laws, California legislation regulates social media by focusing on the design features of these platforms. SB 976, the Protecting Our Kids from Social Media Addiction Act, signed into law by Governor Newsom in September 2024, sets restrictions on operators of any “addictive internet-based service or application” in how they display content and notifications to users under age eighteen.Footnote 167 The act includes social media platforms in their definition of “addictive internet-based service or application,” or a service that “offers or provides users with an addictive feed as a significant part of the service” and further defines “addictive feed” as such service or application “in which multiple pieces of media generated or shared by users are recommended, selected, or prioritized for display to a user based on information provided by the user, or otherwise associated with the user or the user’s device”.Footnote 168 The law imposes several restrictions when the platform has actual knowledge that a user is a minor, including a prohibition on addictive feeds and a restriction on notifications at night and during school hours.Footnote 169 The law also requires platforms to provide mechanisms for parents of minors to set custom no-notification hours and daily time limits; limit their child from viewing the number of likes or other forms of feedback; and set their child’s account to private mode.Footnote 170 The law further requires that platforms set default restrictions on these features for minor accounts, but allows for any restriction to be lifted with parental consent.Footnote 171
NetChoice sued for a preliminary injunction to enjoin the law, which is set to go into effect on January 1, 2027, bringing First Amendment challenges.Footnote 172 The District Court granted in part and denied in part NetChoice’s motion, finding persuasive NetChoice’s argument that prohibiting push notifications at certain times of day and allowing parents to further set those times was likely unconstitutional on the grounds that notifications are expressive speech (under the standard set in Moody v. NetChoice).Footnote 173 However, the Court also found that NetChoice lacked associational standing and lacked ripeness on certain claims, denying the injunction as to those provisions.Footnote 174 On appeal, the Ninth Circuit upheld the District Court’s finding that the likely unconstitutional provisions were severable, and only reversed on one provision of the act, ruling that the prohibition on displaying number of likes or other forms of feedback was unconstitutional as applied because it was not the least restrictive means of achieving California’s goal under strict scrutiny, which applied because it was found to be a content-based restriction.Footnote 175 Despite this reversal, SB 976 largely stands intact, though the question of whether algorithmic social media feeds are expressive speech subject to First Amendment protections is still open. With less than a year before SB 976 goes into effect, the ultimate success of California’s approach remains to be seen, but the Ninth Circuit’s ruling in September was a win for proponents of regulating addictive design features.
C. Litigation
1. State Attorneys General Actions
In 2023, forty-two attorneys general filed a joint lawsuit against Meta.Footnote 176 Thirty-three of these cases, along with hundreds of claims on behalf of children, school districts, and local government entities were consolidated into multidistrict litigation (MDL) in the Northern District of California.Footnote 177 Their complaint alleges that Meta violated numerous state consumer protection laws and federal online privacy law, namely the Children’s Online Privacy Protection Act (COPPA).Footnote 178 COPPA, passed in 1998, imposes requirements on websites and online services that are directed to children under thirteen, or that have actual knowledge that they are collecting personal information online from a child under thirteen.Footnote 179 The complaint also alleged that Meta intentionally designed features to exploit young users while misleading the public about the harms of their platform.Footnote 180
Meta argued that Section 230 of the Communications Decency Act immunized them against many of the claims, but the District Court denied their motion to dismiss on this argument, as the complaint was not about third-party content posted on Meta’s platforms, but rather about Meta’s practices.Footnote 181 In the oral argument on appeal, the Ninth Circuit signaled that they think Section 230 only provides immunity from liability, but not from suit altogether.Footnote 182 The Ninth Circuit has yet to issue a decision, but if it affirms the lower court’s denial, it would be a significant step towards holding tech companies accountable.
A similar case is proceeding in Nevada, in which the state of Nevada made a negligence claim against Snap, Inc., the parent company of Snapchat, under a state deceptive trade practices law.Footnote 183 Snap moved to dismiss the case and brought First Amendment and Section 230 defenses, though the District Court denied that federal authority precluded the state’s negligence claim.Footnote 184 The Supreme Court of Nevada affirmed the District Court’s decision, holding that the State’s claims — that Snap knew it was inflicting harm on young users through the app’s negligent design — did not implicate the First Amendment because the State was not looking to compel speech or place restrictions on editorial functions.Footnote 185 Regarding Section 230, the Court found that the State did not seek liability “for third-party content—thereby taking the underlying complaints outside of Section 230 immunity.”Footnote 186 In allowing the state’s case against Snap to proceed, the Supreme Court of Nevada has reasoned that negligence claims against social media companies can be sufficiently plead without conflicting with the First Amendment or Section 230, providing a workable roadmap for other states seeking accountability.
Another Meta case brought in 2023 by the New Mexico Attorney General concluded in late March 2026 when a jury found that Meta misled users about platform safety and enabled sexual exploitation of minors.Footnote 187 As in the above-mentioned cases, Section 230 did not help Meta because the claims were about Meta’s product design rather than third-party content hosted on the platform.Footnote 188 Raúl Torrez, New Mexico’s Attorney General, secured $375 million in damages under the state’s consumer protection laws and emphasized Meta’s explicit role in causing harm: “Meta executives knew their products harmed children, disregarded warnings from their own employees and lied to the public about what they knew.”Footnote 189 This was the first major loss for Meta amongst a slate of trials set to proceed in 2026, and may present an acute pressure for social media companies to enhance child safety mechanisms on their platforms.
2. Private Tort Claims
In addition to the private claims consolidated into the California MDL, hundreds of claims have been consolidated into a Judicial Counsel Coordination Proceeding (JCCP) in the Superior Court of Los Angeles.Footnote 190 JCCP 5255 represents numerous plaintiffs against Meta, the owner of Instagram and Facebook; ByteDance, which previously was a majority shareholder of TikTok; Snap, the parent company of Snapchat; and Google, as the owner of YouTube.Footnote 191
The bellwether case in JCCP 5255, brought by a plaintiff known as KGM, began jury selection in late January 2026.Footnote 192 KGM is a twenty year-old woman who started using social media at the age of six and alleged that the platforms caused her body dysmorphia and self-harm ideation.Footnote 193 In the week leading up to jury selection, both Snap and TikTok settled with KGM, leaving the claims against Meta and Google to proceed to trial.Footnote 194 In March 2026, a California jury found that Meta and YouTube were negligent in the design of their platforms; knew or should have known that the design was dangerous or likely to be dangerous to minors; and failed to adequately warn of the danger.Footnote 195 The jury awarded KGM three million dollars in compensatory damages and an additional three million dollars in punitive damages, and it agreed that Meta and YouTube “acted with malice, oppression, or fraud.”Footnote 196 While the damages were small relative to the size of Meta and YouTube, this was the first case to go to trial and sets an important precedent for holding accountable social media companies that knowingly caused harm and concealed the dangers of their platforms. The high-profile nature of this case, in addition to the intense scrutiny Meta has already faced from Congress, may put pressure on lawmakers and regulators to find an affirmative solution to protect children and minors going forward.
D. Novel Legal Solutions
Public health regulation has traditionally been left to the states under the Tenth Amendment’s state police power.Footnote 197 States were at the forefront of consumer protections for addiction related to tobacco, alcohol, and gambling, as they are leading the charge today with social media regulation.Footnote 198 Approaches to tobacco, alcohol, and gambling regulations included a strong focus on protection for especially vulnerable populations — children and adolescents.Footnote 199 An approach to social media regulation that is explicitly framed as a public health intervention may help avoid the First Amendment issue of content-based regulations, or it may fit under the Zauderer rational basis standard if a government-mandated warning is either factual and uncontroversial, or fashioned as a remedial measure against deceptive advertising.Footnote 200 This approach is reflected in states that enacted warning label requirements, such as Minnesota, Colorado, New York, and California.Footnote 201 Other areas of law, such as family law, tax law, and privacy law, might provide alternative mechanisms to achieve the same goal of protecting youth from social media harms.Footnote 202
Some jurisdictions have chosen to ban social media entirely for children under a certain age. In 2024, Australia passed a law to ban social media for children under age sixteen, a prohibition that became effective in December 2025 and is currently being enforced on ten platforms: Instagram, Facebook, Snapchat, TikTok, YouTube, X (formerly Twitter), Reddit, Threads, Twitch, and Kick.Footnote 203 The law requires these platforms to take reasonable steps to prevent Australian children under sixteen from holding accounts at the risk of a civil penalty.Footnote 204 In March 2026, Australia’s eSafety Commissioner reported that the proportion of children who had accounts on the implicated platforms decreased from about fifty percent to about thirty-one percent, though when surveying parents whose children had accounts on the platforms prior to the bill’s effective date, around seven in ten reported that their child still had an account on Facebook, Instagram, Snapchat, and TikTok.Footnote 205 Less than half (48.5%) of parents reported that their child still had an account on YouTube.Footnote 206 The report also shared that the Commissioner was undertaking investigations into Meta, Snap, TikTok, and YouTube for failure to comply, which could expose the companies to maximum fines of thirty-four million USD.Footnote 207 While the Australia ban is still in an early stage of implementation, other countries and jurisdictions are considering similar laws to protect minors.
Several countries around the world are considering or have passed social media bans for minors.Footnote 208 Britain, France, Greece, Norway, Poland, and Slovenia are considering bans, and Austria, Indonesia, Malaysia, and Spain have announced or passed prohibitions.Footnote 209 Some bans require or allow for parental consent to grant access to minor users as young as thirteen years old, including those in Brazil, Denmark, Germany, Italy, Norway, and Portugal.Footnote 210 Spain’s Prime Minister described social media as a “failed state … wealthier and more powerful than many nations, including mine.”Footnote 211 It remains to be seen whether these bans will be passed and effectively implemented, but the growing number of countries interested in stricter social media prohibitions reflects an affirmative effort by lawmakers to protect children online.
Alternatively, protecting social media users from addiction is supported by the novel argument for recognition of a fundamental constitutional right to freedom from addiction.Footnote 212 Matthew B. Lawrence, a health law scholar and professor at Emory University School of Law, wrote in 2023 that freedom from addiction has longstanding support in the history and tradition of the United States that “connects the historically marginalized interests of people with substance use and gambling disorders with the increasingly mainstream movement to regulate big tech.”Footnote 213 While it wouldn’t allow regulatory actors complete latitude in their approach to social media companies, recognition of a new fundamental constitutional right would greatly strengthen governmental justifications for legislative efforts at regulation.
V. Proposed Technical Solutions
In addition to legal solutions through regulation and litigation, there are two types of technical solutions: those adopted by companies themselves, and those offered by third parties. Companies are unlikely to affirmatively limit features that have garnered them unprecedented influence, power, and money, and rather need to be pushed towards implementing technical solutions, whether through straightforward motivations such as good PR or punitive assessments such as civil fines.Footnote 214
The most effective design changes — eliminating algorithmic recommendations, infinite scroll, and notifications — are feasible for platforms to carry out in the United States, as Facebook, Instagram, and TikTok are already offering chronological and non-personalized feeds for use in Europe in compliance with the European Union’s Digital Services Act.Footnote 215 Indeed, Instagram offers mobile app users the ability to limit their feed to content from other users that they follow, rather than having it injected with suggested content.Footnote 216 App users also have the option to hide suggested posts or to “snooze” all suggested content for thirty days.Footnote 217 Reducing addictive design is clearly possible. Platforms may simply need adequate incentives to do so.
Litigation incentives can further persuade technical changes. Liability for tort or consumer protection claims turn on whether a platform’s design was reasonable, which requires companies to consider what alternatives exist. Following the bellwether KGM case, which saw a jury verdict for the plaintiff against Meta and YouTube, and settlements with TikTok and Snap, social media companies may think twice about their design features when it comes to minors.Footnote 218 Given that numerous similar cases are set to take the stage in the coming months, the financial damages and the extended press coverage may push companies to reconsider addictive design features and tighter protocols for children and teens. The challenge for companies is how to preserve the benefits of social media while limiting the harms to minors — and the companies’ exposure to lawsuits. Instagram is among the platforms that has already announced one such technical solution aimed at higher protections for adolescents.
A. Instagram Teen Accounts
In September 2024, Meta announced Instagram Teen Accounts, with built-in protections for users under eighteen years of age.Footnote 219 Meta automatically places teens in Teen Accounts, with default settings including private profiles, messaging restrictions, time limit reminders after sixty minutes, and sleep mode from 10 PM to 7 AM, during which notifications are muted.Footnote 220 Parents are given supervision controls for teen accounts under age sixteen, with these users required to obtain parental permission to change any settings.Footnote 221 Parental supervision can be enabled for teen accounts with users older than sixteen but is not automatically granted.Footnote 222 Parental supervision also allows the parent to set time limits, sleep mode hours, send reminders to exit the app, block access to the app, see content topics that their teen indicated interest in, and see users or groups that their teen chats with (though the messages themselves are not accessible to the parent).Footnote 223 As part of this new feature, Instagram is also changing its age verification practices to restrict access for users who attempt to get around the age gate, including proactive measures to identify teens who are operating adult accounts.Footnote 224
It is unclear whether this change, which directly targets some of the features of addictive design that can capture adolescent users, is in response to legislation such as the SAFE for Kids Act.Footnote 225 Alternatively, it could be a strategic business move to improve the company’s image, especially after The Facebook Files investigation revealed the extent of Meta’s knowledge of Instagram’s harms to teenagers.Footnote 226 Furthermore, it remains to be seen whether the introduction of Instagram Teen Accounts will reduce or mitigate the adverse mental health outcomes associated with the platform. Still, it is promising that large companies like Meta are introducing products that grant greater control to parents of adolescent consumers.
B. Digital Detox
In addition to those offered by social media companies, there are also private ordering solutions. These market-based solutions have been attempted by multiple third parties and have generally faced serious objection from platforms.
In 2018, British software developer Louis Barclay developed a browser extension called Nudge.Footnote 227 Nudge is designed to make social media less addictive by using the same principles of behavioral psychology that platforms rely on to facilitate addiction in the first place.Footnote 228 Like many social media platforms, Nudge uses the Hook model, inserting “nudges” at points throughout the feedback loop to break the cycle of reinforcement, and therefore, addiction.Footnote 229
Recall the four stages of the Hook model: trigger, action, reward, and investment.Footnote 230 At the trigger stage, Nudge works using “default nudge” and “hiding nudge.”Footnote 231 The default nudge utilizes an “opt-out” system. Rather than loading Facebook, the user is taken to a default page that asks the user to confirm they would like to proceed to Facebook.Footnote 232 The hiding nudge obscures triggers (notifications, likes, and comments), thereby creating a barrier between the trigger and action stages.Footnote 233
At the action stage, Nudge creates friction.Footnote 234 Just as features like infinite scroll aim to limit friction, Nudge counteracts the seamlessness with which users become sucked in by prompting users to confirm actions, such as viewing suggested content.Footnote 235
In the reward stage, Nudge aims to reduce the rewards of staying online through pause-reminders and “feedback nudge.”Footnote 236 Pause-reminders force the user to consider how long they have been online, making them more likely to log off.Footnote 237 The feedback nudge similarly informs users of how long they’ve spent online, however it incorporates a visual icon on the user’s screen that grows bigger by the minute.Footnote 238
In the final stage of the Hook model, the investment stage, Nudge works to reduce a user’s investment in a platform with the “unfollow nudge.”Footnote 239 The unfollowing process could alternatively be automated, and indeed Barclay did develop an automated unfollow tool, Unfollow Everything, in July 2020.Footnote 240 This browser extension allowed users to delete their Facebook news feed by unfollowing all friends, groups, and pages, while maintaining group memberships and friends.Footnote 241 Unfollow Everything allowed users to use Facebook without the behavioral temptation of an infinite scroll news feed, or to carefully curate the news feed by refollowing select friends, groups, and pages.Footnote 242
A major barrier to browser extensions and tools such as Nudge and Unfollow Everything is the pushback from social media platforms. Unfollow Everything is no longer available to Facebook users, and Facebook banned Barclay for life.Footnote 243 In 2021, Barclay wrote:
Facebook’s behavior isn’t just anti-competitive; it’s anti-consumer. We are being locked into platforms by virtue of their undeniable usefulness, and then prevented from making legitimate choices over how we use them—not just through the squashing of tools like Unfollow Everything, but through the highly manipulative designs and features platforms adopt in the first place. The loser here is the user, and the cost is counted in billions of wasted hours spent on Facebook.Footnote 244
One problem with these consumer tools is that they require affirmative action on the part of the user — that is, a user must seek to reduce the effects of social media in their lives to find, use, and benefit from such tools, which adolescents may be less likely to do. Further, until social media platforms and big tech companies become more friendly towards consumer tools that aim to reduce the effects of addictive design, adolescent users may need to find solutions elsewhere.
VI. Conclusion
The evidence presented throughout this paper establishes a compelling case for regulatory intervention to protect adolescents from the addictive qualities of social media platforms. The neurological similarities between problematic social media use and substance use disorders, combined with platforms’ deliberate exploitation of psychological vulnerabilities through addictive design features, demonstrate that this is fundamentally a public health issue.
Looking ahead, a comprehensive regulatory approach should draw lessons from public health interventions for tobacco, alcohol, and gambling, especially given the similar neurological and behavioral pathways that social media platforms rely on. Legal approaches focused on regulating addictive design rather than moderating content offer the most promising path forward. By targeting structural design choices that facilitate addiction — such as algorithmic feeds, notification systems, and infinite scroll — regulators can avoid the First Amendment challenges that have derailed previous state-level attempts at social media regulation. New York’s SAFE for Kids Act represents a significant step in this direction, addressing the mechanisms of addiction without directly regulating content.
Technical solutions also show promise, particularly when implemented as default settings rather than opt-in features. Meta’s introduction of Instagram Teen Accounts with built-in protections demonstrates that platforms have the capability to reduce addictive features when sufficiently motivated, however it is likely that platforms will require external legal pressure or increased public pressure to make meaningful changes. This pressure might come in the form of the numerous ongoing cases against companies like Meta, YouTube, Snap, and TikTok. On the heels of a bellwether victory for a twenty-year-old plaintiff in California and the success of the New Mexico Attorney General’s negligence suit against Meta, social media companies may consider whether alternative design choices are advisable. On the international stage, multiple countries have implemented or are in the process of enacting social media bans. Though the efficacy of this approach has yet to be proven, companies should take notice of the growing negative sentiment towards social media influence on adolescent users and the increased political will to act in order to achieve prohibition and accountability.
Social media will only become more embedded in the everyday lives of adolescents in the United States, increasing the potential for lifelong impacts on users’ mental health. By framing social media addiction as a public health issue and implementing evidence-based regulations that target addictive design choices rather than content, lawmakers and regulators can protect vulnerable young users while preserving the benefits of digital connectivity, meanwhile past and ongoing harms can be addressed through the courts. This balanced approach recognizes that social media, like other addictive products, requires thoughtful regulation to ensure that its tremendous benefits do not come at the expense of our children’s wellbeing.
Open access
