What the Platforms Are Doing to Our Children

Your daughter is fourteen. She goes to bed at ten. You check on her at midnight and she's still awake, phone face-up in the dark, scrolling. You tell yourself it's a phase. You tell yourself she's just connecting with friends. You tell yourself the world is different now, kids are different now, and this is just how it is.

It isn't just how it is. Someone built it this way on purpose.

The mental health numbers for American adolescents tell a story that ought to stop every parent cold. In 2011, about 18 percent of teenage girls reported high levels of depression. By 2021, that number was 40 percent. By 2023, 57 percent of girls exhibited symptoms of depression, compared to 31 percent of boys (CDC, 2023). Emergency room admissions for self-harm among girls ages 10 to 14 tripled between 2009 and 2015 (CDC, 2017). The CDC found that nearly one in three high school girls considered suicide in 2021, a 60 percent increase since 2011 (CDC, 2022).

These are not small numbers. A doubling of depression rates in a single decade is not a statistical fluctuation. It is a signal.

The phones arrived around 2012. The social media platforms followed. The timing is not coincidental.

Jonathan Haidt, a social psychologist at New York University's Stern School of Business, spent years assembling the evidence in "The Anxious Generation," published in 2024. His central claim: when adolescents' social lives migrated from the physical world onto smartphones and social media platforms, anxiety and depression followed. The number of adolescents reporting depression rose 134 percent between 2012 and 2020. Anxiety rose 106 percent (Haidt, "The Anxious Generation," 2024). The pattern holds across the United States, the United Kingdom, Canada, and Australia, suggesting something systemic rather than local.

The book has critics. Candice Odgers, a developmental psychologist at UC Irvine, published a review in Nature in 2024 arguing that most empirical research does not find large or consistently negative effects from social media on adolescent mental health (Odgers, Nature, 2024). Her point deserves to be taken seriously. Correlation is not causation. Longitudinal studies are hard. Kids who are already struggling may use social media more, which would reverse the causal arrow.

But here is what the critique cannot explain away: the internal research the platforms conducted themselves.

In 2019, Facebook commissioned its own research on how Instagram affects teenage girls. Their researchers found that Instagram was making body image issues worse for one in three teen girls. An internal presentation slide read: "We make body image issues worse for one in three teen girls." Among teens who reported suicidal thoughts, 13 percent of British users and 6 percent of American users traced the feeling back to the platform (Wells, Horwitz, and Seetharaman, The Wall Street Journal, September 14, 2021). Facebook knew. It continued anyway.

This is not a debate about methodology. This is internal evidence that the company's own researchers found harm, documented it, and the company kept building. The Wall Street Journal published these findings in September 2021 as part of what became known as the Facebook Files. Facebook never made the research public or available to academics or lawmakers who requested it (WSJ, 2021).

The mechanism matters. Social media platforms are optimized for engagement, which means they are optimized for the emotions that drive engagement: comparison, anxiety, fear, outrage, envy. A teenage girl scrolling Instagram is not consuming a neutral feed. She is encountering a machine calibrated to trigger emotional responses that keep her on the platform longer. Her discomfort is, by the platform's logic, a feature. Her addiction is revenue.

The U.S. Surgeon General, Dr. Vivek Murthy, issued an advisory in May 2023 that documented what the research actually shows. Up to 95 percent of youth ages 13 to 17 report using a social media platform. More than a third say they use it "almost constantly." The average daily screen time for American teens is 8.5 hours, of which 3.5 hours is on social media alone (Surgeon General's Advisory, HHS, 2023). Teens who use social media for more than three hours per day face double the risk of depression and anxiety symptoms, and the average teen is already above that threshold (Surgeon General's Advisory, HHS, 2023). Murthy concluded that there is not yet enough evidence to determine whether social media is sufficiently safe for children.

By June 2024, Murthy was calling for warning labels on social media platforms, the same kind that appear on cigarettes and alcohol (Murthy, The New York Times, June 2024). He framed it plainly: "The platforms are not just failing to protect our children. They are amplifying the risks."

The American Psychological Association issued its own health advisory in April 2023. Among its ten recommendations: limit social media use so it does not interfere with sleep or physical activity; minimize exposure to content that promotes self-harm, disordered eating, or social comparison; ensure children possess psychological maturity before using social media (APA Health Advisory on Social Media Use in Adolescence, 2023). These are the recommendations of the nation's largest professional psychology organization, and they read like a list of things the platforms were designed to undermine.

None of this is paranoia. It is the ordinary accounting of how attention-economy businesses work. Their revenue depends on time-on-platform. Time-on-platform depends on emotional activation. Adolescents are particularly vulnerable. Their prefrontal cortex, the part of the brain that governs self-regulation, is still forming. An algorithm trained on behavioral data will find, for each individual user, the content that provokes the sharpest response. For a teenage girl, that content is usually about appearance, status, and belonging. The algorithm does not care what it finds. It optimizes. The product works as designed.

The congressional hearings, the litigation, the leaked documents: none of it changed the business model. Meta and Alphabet combined for over $425 billion in advertising revenue in 2024, built on behavioral data harvested from every click and scroll (Alphabet Inc., 2025; Meta Platforms, 2025). The architecture that harms children is also the architecture that generates returns. A company cannot reform that from the inside without ceasing to be that company.

That points to a harder question: what did we lose when childhood moved indoors and online, and is any of it recoverable?

Haidt's argument is not simply that phones are dangerous. It's that the phone-based childhood replaced a different kind of childhood, one organized around physical risk, unstructured time, face-to-face failure, and boredom. The play-based childhood was not idyllic. Kids got hurt. They were bored. They worked out social conflicts without adult mediation and sometimes got it wrong. But those were the conditions under which certain things developed: frustration tolerance, a sense of physical competence, the ability to read a room, the capacity to sit alone without reaching for something (Haidt, "The Anxious Generation," 2024).

The brain arrives at adolescence primed for social comparison. This is not a design flaw. It is how children have always calibrated their place in a community, learned which behaviors are valued, figured out who they are in relation to others. For most of human history, that comparison was local and bounded. You knew the thirty or forty kids in your school. You could read their faces. The feedback was real and contextual and it ended when you went home.

Instagram made the reference group global and permanent. A girl measuring herself against a curated feed of thousands of strangers, each image selected for maximum social impact, each one representing a highlight reel rather than a life, is not doing what the brain evolved to do with social comparison. She is running a process designed for a village on inputs from a stadium. The result is predictable (Twenge, iGen, 2017).

The technology critique that matters is not about screen time as a quantity. It is about the specific mechanics of how these platforms are designed. Infinite scroll removes the natural stopping point that bounded media always had. Variable reward schedules (sometimes you get likes, sometimes you don't, and the uncertainty is the point) are the same mechanism that makes slot machines hard to leave. Algorithmic amplification means the content that provokes the strongest emotional response gets shown more. None of these features are accidental. Each one was chosen because it increases time-on-platform. Each one is incompatible with a child's developmental needs.

That shapes what a genuine alternative would have to look like. Not a safer version of the same product, with better parental controls and a kinder algorithm. The engagement-maximization model is the problem. Any technology built around that model, however tasteful its branding, will trend toward the same outcomes. The question is whether a different economic model for children's technology is viable, one where the product succeeds by being genuinely useful and then getting out of the way, rather than by capturing as many hours as possible.

We do not have a settled answer. What we have is a generation of children whose mental health data tells us clearly that the current model is not working, a handful of researchers documenting what was lost, and a set of design questions that have barely been asked. What does it mean to build technology that treats the successful outcome as a child who used the tool and then went outside? What does communication software look like when the success metric is depth of connection rather than frequency of engagement? What would a social platform feel like if it showed you fewer posts, not more, and considered restraint a feature?

Haidt's four proposals (no smartphones before high school, no social media before 16, phone-free schools, and real-world independence for children) are not primarily technology proposals (Haidt, "The Anxious Generation," 2024). They are proposals about what childhood should be protected for. The technology question comes after: once you have decided that a child's development requires boredom and physical risk and face-to-face relationships, you can start asking what tools would support that rather than erode it.

Those tools do not yet exist at any meaningful scale. The market for children's technology has been defined almost entirely by the engagement model. Building outside that model requires believing the alternative is possible before the alternative exists. That has always been the uncomfortable position. Most necessary things start there.

---

Sources

Haidt, Jonathan. "The Anxious Generation: How the Great Rewiring of Childhood Is Causing an Epidemic of Mental Illness." Penguin Press, 2024.

Odgers, Candice. "The Great Rewiring: Is Social Media Really Behind an Epidemic of Teenage Mental Illness?" Nature, March 2024. https://www.nature.com/articles/d41586-024-00902-2

Wells, Georgia, Jeff Horwitz, and Deepa Seetharaman. "Facebook Knows Instagram Is Toxic for Teen Girls, Company Documents Show." The Wall Street Journal, September 14, 2021. https://www.wsj.com/articles/facebook-knows-instagram-is-toxic-for-teen-girls-company-documents-show-11631620739

Murthy, Vivek H. "Social Media and Youth Mental Health: The U.S. Surgeon General's Advisory." U.S. Department of Health and Human Services, May 2023. https://www.hhs.gov/sites/default/files/sg-youth-mental-health-social-media-advisory.pdf

Murthy, Vivek H. "Surgeon General: Why I'm Calling for a Warning Label on Social Media Platforms." The New York Times, June 17, 2024.

American Psychological Association. "Health Advisory on Social Media Use in Adolescence." APA, April 2023. https://www.apa.org/topics/social-media-internet/health-advisory-adolescent-social-media-use

Centers for Disease Control and Prevention. "Youth Risk Behavior Surveillance Data Summary and Trends Report: 2011–2021." CDC, 2022. https://www.cdc.gov/healthyyouth/data/yrbs/pdf/YRBS_Data-Summary-Trends_Report2023_508.pdf

Centers for Disease Control and Prevention. "Emergency Department Visit Rates for Self-Harm Among Young People, 2009–2015." NCHS Data Brief, 2017.

Centers for Disease Control and Prevention. "Youth Mental Health: The Numbers." CDC, 2023. https://www.cdc.gov/healthy-youth/mental-health/mental-health-numbers.html

Alphabet Inc. "Alphabet Announces Fourth Quarter and Fiscal Year 2024 Results." Investor Relations, January 2025. https://abc.xyz/investor/

Meta Platforms, Inc. "Meta Reports Fourth Quarter and Full Year 2024 Results." Investor Relations, January 2025. https://investor.atmeta.com/investor-news/press-release-details/2025/Meta-Reports-Fourth-Quarter-and-Full-Year-2024-Results/

Twenge, Jean M. "iGen: Why Today's Super-Connected Kids Are Growing Up Less Rebellious, More Tolerant, Less Happy — and Completely Unprepared for Adulthood." Atria Books, 2017.