Every week there's a new headline warning that artificial intelligence is coming for your job. Executives quote it in earnings calls, LinkedIn is full of it, and honestly, it's a legitimate thing to worry about. But while everyone's been staring at that particular horizon, something far more immediate has been happening in the next room — on the phone in your teenager's hand, at your dinner table, in your own pocket right now.
We need to talk about what short-form video apps are actually doing to people, because the honest comparison is not close. AI, so far, is mostly a threat to how we earn a living. Reels, TikTok, and their imitators are a documented threat to how young brains develop, how teenagers see themselves, and in a growing number of very real, very reported cases, to whether they come home at all.
Let's deal with the AI panic first, because the data doesn't support the doom
The World Economic Forum's Future of Jobs Report 2025 projects that AI and related shifts will displace about 92 million jobs globally by 2030 — a real number, and real disruption for the people in those roles, particularly cashiers and administrative work. But the same report projects 170 million new roles being created in the same window, for a net gain of 78 million jobs. Roughly 22% of today's jobs will be touched by this shift, and 59 out of every 100 workers will need reskilling, which is its own serious challenge.
That's a real economic transition, and it deserves real policy attention. But notice what it isn't: it isn't children dying. It isn't a documented, growing body of clinical research linking a product directly to anxiety, depression, and self-harm in kids as young as ten. AI job disruption is a slow-moving, largely reversible economic problem. What's happening with short-form video is neither slow nor, for the families affected, reversible.
What these apps are actually doing to attention and the developing brain
This isn't a vibes-based claim. A 2025 systematic review published on medRxiv looked at 17 peer-reviewed studies on short-form video (SFV) use among Gen Z adults and found a "consistent association with attentional disruption" — fragmented focus, difficulty sustaining concentration, and a reduced ability to filter out distractions. The same review found impairments in short-term and prospective memory when video use competed with learning tasks, plus elevated anxiety linked to heavy use and a pattern of compulsive, hard-to-stop scrolling.
The neuroscience is starting to catch up to the behavioral data. Researchers have begun using brain imaging to look directly at what happens during short-video exposure — a 2026 study in the journal Neuroscience Bulletin examined dopamine-sensitive striatal connectivity during short-video viewing and its relationship to addictive-like engagement traits, part of a fast-growing body of neuroimaging work specifically probing what researchers have started calling "TikTok brain." None of this is settled science yet, and researchers are appropriately cautious about causation. But the direction of the evidence, across behavioral studies and now brain imaging, points the same way.
Here's why that happens, mechanically: these apps are built on an endless, algorithmically-optimized feed with no natural stopping point, where every swipe has a chance of delivering something more interesting than the last thing — the same variable-reward structure that makes slot machines compelling. Your brain doesn't get a satisfying, complete experience and move on. It gets a hit of novelty every six to fifteen seconds, and the app is engineered, A/B tested, and refined specifically to keep that cycle going as long as possible. A long-form video or a book asks your brain to sustain attention. A Reels feed trains it not to.
The apps we're actually talking about
This isn't a problem with one app. It's a business model that most major platforms have converged on:
- TikTok — the one that built the format and remains the most studied and most litigated
- Instagram Reels (Meta)
- YouTube Shorts (Google)
- Snapchat Spotlight
- Facebook — both its main feed and its own Reels product
- Triller and a wave of smaller TikTok clones
- Livestreaming platforms like Kick and Twitch, which add a different but related danger: real-time audiences egging people on in the moment
Different logos, same underlying mechanism: infinite scroll, autoplay, and a recommendation algorithm optimized for one metric — time on app — with a young, developing, highly persuadable user base. Notably, these are the exact same ten platforms — Instagram, Facebook, Threads, Snapchat, TikTok, YouTube, X, Reddit, Twitch, and Kick — that Australia named specifically when it banned under-16 access nationwide (more on that below).
The platforms' own research says the quiet part out loud
You don't have to take an outside critic's word for this. In 2021, the Wall Street Journal published what became known as the Facebook Files — Meta's own internal research on Instagram's effect on teen users. The findings, later confirmed in congressional testimony, were stark: 32% of teen girls said that when they felt bad about their bodies, Instagram made them feel worse. Among teens who reported suicidal thoughts, 13% of British users and 6% of American users traced those thoughts specifically to Instagram. About 14% of teen boys said the app negatively affected how they saw themselves. Meta's own researchers wrote that "aspects of Instagram exacerbate each other to create a perfect storm."
That's not an outside allegation. That's the company's own data.
The U.S. Surgeon General took this seriously enough to issue a formal 2023 advisory on social media and youth mental health, and more recently called for warning labels on social media platforms altogether — the same kind of regulatory step this country has historically reserved for cigarettes. The American Psychological Association's 2023 Health Advisory on Social Media Use in Adolescence reached a similar conclusion from the clinical side, recommending active adult monitoring during early adolescence, limits on exposure to content that normalizes self-harm, and specific screening for signs of problematic, compulsive use. NYU social psychologist Jonathan Haidt built an entire, extensively footnoted book, The Anxious Generation, around what he calls "the great rewiring of childhood," arguing that the shift from free play to smartphone-and-social-media childhoods lines up almost exactly with the timeline of a global rise in teen anxiety and depression.
How big is this, really? The numbers behind the panic
Pew Research Center's most recent teen survey data, published in 2025, puts real numbers on how total the saturation is. 73% of teens say they use YouTube daily, including 15% who describe their use as almost constant. A majority visit TikTok daily too, with 16% reporting near-constant use. Around half use Instagram or Snapchat daily, with roughly one in ten on each almost constantly.
What's changed is that teens themselves are starting to notice. 45% of teens now say they spend too much time on social media, up sharply from 27% just two years earlier in 2023. 44% say they've actually tried to cut back, versus 39% in 2023 — and yet 55% say they haven't managed to reduce their use at all. That gap, between wanting to stop and being unable to, is worth sitting with. It's not a description of a hobby. It's a description of something a lot of teenagers already recognize, in their own words, as a habit they can't fully control.
When "for the views" becomes a death sentence
This is the part that should stop the conversation cold, because it isn't abstract, and it isn't rare enough to dismiss as a freak accident.
The "blackout challenge" — a self-choking stunt that went viral on TikTok starting in 2021 — has been linked by Bloomberg Businessweek's reporting to roughly 20 child deaths in an 18-month span alone, part of a documented toll that outlets and court filings put between 15 and 20-plus confirmed fatalities, with some estimates higher. It's serious enough that in August 2024, the U.S. Third Circuit Court of Appeals reversed a lower court's ruling and said TikTok could actually be held legally accountable over how its algorithm promoted the challenge to children, and in February 2025 the parents of four British teenagers who died attempting it filed suit against the platform.
That's one challenge. A separate analysis covered by Vice found that the "Benadryl challenge" — where teens film themselves overdosing on allergy medication to hallucinate — was directly tied to the 2023 death of 13-year-old Jacob Stevens in Ohio. The "Tide Pod Challenge" sent more than 35,000 under-18s to emergency rooms between 2016 and 2020 for chemical burns and seizures. The "Milk Crate Challenge" sent over 8,000 people to the ER, with injuries ranging from fractures to spinal damage and paralysis.
And it isn't only staged "challenges." It's people filming ordinary content who die chasing the shot: a 27-year-old woman in India died filming a staged hanging scene for a reel. A student in Himachal Pradesh died attempting a bike stunt for a reel. A boy was struck and killed by a train while filming a video for social media. A social media influencer fell to his death mid-stunt during a publicity shoot. A Mexican influencer was shot and killed on a livestream, in front of an audience watching in real time. None of these are urban legends. Every one of them is a reported news story, and every one of them was, in some direct way, performed for an audience that was never going to do anything but watch.
The scale: this isn't a fringe problem
According to CDC survey data covering July 2021 through December 2023, just over 50% of American teenagers report 4 or more hours of screen time a day, rising to 55% among 15-to-17-year-olds. Among teens with 4+ hours of daily screen time, roughly 1 in 4 reported symptoms of anxiety (27.1%) or depression (25.9%) — roughly double the rate seen in teens with lower screen time. That's not proof that screens alone cause anxiety and depression, and researchers are careful about overclaiming causation from correlational data. But it's a big, consistent, government-collected signal that something about how young people are spending four-plus hours a day deserves more scrutiny than it's currently getting.
To be fair, for a second
It would be dishonest to pretend social media is purely destructive. These platforms have also given teenagers community, creative outlets, and connection, especially for kids who are isolated, LGBTQ+, disabled, or otherwise underserved by their immediate surroundings. Most of the research in this space is correlational, not experimental, and researchers are explicit that causation is hard to prove from self-report and cross-sectional data. This isn't a call to ban technology or pretend every teenager on TikTok is in danger. It's a call to be honest about a documented pattern of harm that the platforms' own research has already confirmed, and to stop treating "the algorithm made my kid want to hurt themselves for views" as a fringe concern when it's now sitting in court filings, Surgeon General advisories, and national legislation.
What we can actually do about it
None of this is hopeless. There are concrete steps that work, at both the household and policy level.
- Delay the smartphone, not just the app. Haidt's research and multiple pediatric groups recommend holding off on smartphones and social media accounts until at least age 14–16, using a basic phone or watch for contact in the meantime.
- Use the built-in tools you already have. Apple's Screen Time and Google's Family Link both let you set hard daily limits per app, schedule downtime, and require approval for new app installs — free, and already on the device.
- Add a dedicated layer if you need more. Third-party tools like Bark, Qustodio, and Boomerang go further, with content monitoring and app-specific time limits across both iOS and Android.
- Turn off autoplay and disable the algorithmic feed where the platform allows it. Several apps now let you switch to a chronological, "following-only" feed, which removes the engagement-optimized recommendation engine that drives compulsive use.
- Make phones leave the bedroom at night. A charging station outside the bedroom is one of the single highest-leverage changes a household can make, because it fixes sleep, which fixes almost everything else.
- Talk about specific challenges directly, not just "be careful online." Vague warnings don't work. Naming the blackout challenge, the Benadryl challenge, and similar trends by name, and explaining exactly why they're dangerous, works better than generic caution.
- Support the policy layer, because individual parenting can't fully solve this alone. Australia's under-16 social media ban, which took effect on December 10, 2025, is the clearest real-world test case so far: it covers ten platforms by name (Instagram, Facebook, Threads, Snapchat, TikTok, YouTube, X, Reddit, Twitch, and Kick), requires companies to take "reasonable steps" to verify age, and carries fines of up to roughly A$49.5 million (about $32 million USD) for platforms that don't comply. Whatever you think of the specific mechanism, it's the first serious attempt by a national government to shift the burden off individual families, and other countries are watching closely to see if it works.
The bottom line
AI might restructure how some of us earn a living over the next decade, and that's a real conversation worth having, seriously and without panic. But it is not, today, sending teenagers to the emergency room in the tens of thousands, and it is not the subject of a company's own internal research admitting it makes a third of teen girls feel worse about their bodies. The apps already in every teenager's pocket are doing both of those things right now, documented, sourced, and in court. That's the threat that actually deserves the moral panic everyone keeps aiming at AI.
Sources
- World Economic Forum — Future of Jobs Report 2025
- medRxiv — systematic review on short-form video and cognition (17 studies, 2025)
- Neuroscience Bulletin (ScienceDirect) — dopamine-sensitive striatal connectivity during short-video exposure, 2026
- CNBC / Wall Street Journal — "Facebook Files," 2021
- U.S. Surgeon General — Advisory on Social Media and Youth Mental Health, 2023
- American Psychological Association — Health Advisory on Social Media Use in Adolescence, 2023
- Jonathan Haidt — The Anxious Generation
- Pew Research Center — "10 facts about teens and social media," 2025
- Wikipedia / Bloomberg Businessweek reporting — the blackout challenge
- Vice — "5 TikTok Challenges That Left Thousands Injured and Over 100 Dead"
- CDC NCHS Data Brief 513 — "Daily Screen Time Among Teenagers," 2024
- TIME — Australia's under-16 social media ban, explained, 2025
