What happens on social media doesn't stay on social media.
In the space between a scrolling thumb and a clenched fist, a culture is being quietly remade. Researchers and scholars are documenting what many have long suspected: the misogynistic content now flowing freely through mainstream social media platforms is not merely offensive speech, but a conditioning force — one that correlates measurably with rising rates of sexual violence, eroding policy protections, and the deaths of women who could not escape the hatred that followed them offline. The digital and physical worlds, it turns out, have never been separate; they are mirrors, and what one reflects, the other enacts.
- TikTok's algorithm served up misogynistic content at four times the rate within just five days of mimicking a young man's browsing habits — not a malfunction, but the platform performing exactly as engineered.
- Extreme misogyny once confined to fringe forums has colonized mainstream platforms, its edges softened by irony and humor until the shock dissolves and the normalization takes hold.
- The human cost is no longer abstract: an estimated 154,000 additional high school girls experienced sexual coercion in a single decade, women died by suicide after online pile-ons, and domestic violence surged globally as pandemic isolation fused with digital hatred.
- Policy has followed culture — abortion rights overturned, childcare funding expired, nutrition programs gutted — as if lawmakers sensed that a society primed to accept misogyny online would accept its consequences in law.
- Advocates are pushing back on multiple fronts: demanding serious content moderation from tech companies, urging users to report violations, and calling for offline conversations that interrupt radicalization before it compounds.
When a researcher at University College London created accounts mimicking the interests of lonely young men and let TikTok's algorithm run, misogynistic content quadrupled within five days. It was not a bug. It was the system doing what it was built to do — keep users engaged, whatever the cost.
Kaitlyn Regehr, who studies online extremist groups, describes this as a watershed. The virulent misogyny that once festered in obscure forums has migrated to platforms used by hundreds of millions, arriving wrapped in edgy humor until it no longer registers as extreme. The normalization, she argues, is now complete.
The offline world is absorbing the lesson. CDC data shows the share of US high school girls reporting forced sex rose from 12 to 14 percent between 2011 and 2021 — a two-point shift that translates to roughly 154,000 additional girls subjected to sexual coercion. Research consistently links consumption of violent media to real-world violence. The algorithm doesn't merely reflect culture; it instructs it.
The pattern reaches further. During the pandemic, women in Africa faced a surge in killings as domestic violence spiked globally. In China, a 23-year-old woman died by suicide after strangers attacked her online for sharing news of her graduate school acceptance. A high school teacher suffered a fatal heart attack after hackers disrupted her classes. The harassment was digital. The deaths were not.
Policy, too, has shifted in ways that seem to track the cultural current. Within a year of Roe v. Wade being overturned, half of US states had restricted or banned abortion. Congress allowed pandemic-era childcare funding to expire — a move projected to shutter over 70,000 programs and displace 3.2 million children. Nutrition support for pregnant women and young children lost critical funding. Two million women and children will go without.
The question scholars like Kara Alaimo keep fielding — why don't women simply leave the platforms? — misses the point entirely. Deleting an account does not stop the algorithm from radicalizing someone else. The solution demands action at every level: tech companies must treat content moderation as a genuine obligation, users must report violations and flood feeds with content that affirms women's dignity, and those who encounter misogyny among friends and family must be willing to have the harder conversation offline. What circulates on social media does not stay there. It never did.
The connection between what happens on a screen and what happens in the world is no longer theoretical. When a researcher at University College London set up accounts mimicking the interests of young men—loneliness, mental health, fitness—and scrolled through TikTok, the platform's algorithm served up misogynistic content at a rate that quadrupled in five days. This wasn't a glitch. It was the system working as designed, surfacing material that kept users engaged, regardless of what that material was teaching them about women.
Kaitlyn Regehr, who studies online extremist groups, calls this a watershed moment. The kind of extreme misogyny that once lived in the darker corners of the internet—in forums and message boards where men gathered to nurse grievances against women—has migrated to platforms with hundreds of millions of users. What would have seemed shocking to post on mainstream social networks a few years ago now circulates freely, wrapped in the language of edgy humor. The shock has worn off. The normalization is complete.
This matters because the offline world is listening. In 2021, the US Centers for Disease Control released data showing that the percentage of high school girls reporting forced sex had climbed from 12 percent in 2011 to 14 percent by 2021. With roughly 15.4 million high school students in the country and half of them female, that two-point increase translates to approximately 154,000 additional girls subjected to sexual coercion in a single decade. Research consistently shows that people who consume depictions of violence are more likely to commit violence themselves. The algorithm doesn't just show you what exists; it teaches you what's acceptable.
The pattern extends beyond sexual assault. During the pandemic, when everyone retreated online, women in Africa reported a surge in killings. Domestic violence spiked as women found themselves trapped in homes with abusive partners, their screens their only window to the world—a world that was simultaneously telling them they deserved what was happening. In China, a 23-year-old woman took her own life after being attacked on social media for visiting her grandfather in the hospital and sharing news of her graduate school admission. Strangers on the internet called her a prostitute because of her hair color. Another woman, a high school teacher, suffered a fatal heart attack after hackers disrupted her online classes, preventing her from teaching.
The erosion extends to policy. By June 2023, a year after the Supreme Court overturned Roe v. Wade, half of all US states had passed laws banning or restricting abortion. For the court to believe this reversal was politically feasible, it had to assume women would accept the loss of a potentially life-saving right. One factor in that calculation may have been the cultural shift—extremely online, extremely visible—toward accepting and even celebrating misogyny. That same year, Congress allowed billions in pandemic-era childcare funding to expire, a decision the Century Foundation estimates will close over 70,000 childcare programs and leave 3.2 million children without care. Millions of mothers will be forced to leave the workforce or reduce their hours. Simultaneously, the nutrition program for pregnant women, postpartum women, and young children lost funding it desperately needed as participation and food costs rose. Two million women and children will go without the help they need to eat.
The question Kara Alaimo, an associate professor of communication who has written extensively on this subject, keeps hearing is simple: Why don't women just delete the apps? The answer is equally simple: because other people are using those platforms to spread the hatred that is making women less safe. Deleting your account doesn't stop the algorithm from radicalizing someone else. It doesn't stop the normalization. The solution requires action at every level. Tech companies need to stop hosting misogynistic content and get serious about removing it, using both human moderators and artificial intelligence. When you see friends or family members posting it, you need to have a conversation offline—what activists call "calling them in"—rather than engaging with trolls publicly, which only amplifies their reach. Report violations to the platforms. And crucially, flood the feeds with content that empowers women, content that celebrates women's power and accomplishment, so the algorithm learns to serve that instead.
What happens on social media doesn't stay on social media. The violence celebrated in comments sections becomes violence in bedrooms and streets. The rights stripped away in policy chambers are stripped away because a culture has been primed to accept it. The two worlds are not separate. They never were.
Citas Notables
Extremist misogyny that was once segregated to niche platforms is now disseminating onto much more popular platforms like TikTok and permeating into youth culture more generally.— Kaitlyn Regehr, associate professor at University College London who studies online extremist groups
We are the first generation in American history to have to tell the next generation they have less rights than us.— Linda Sarsour, Palestinian-American feminist activist
La Conversación del Hearth Otra perspectiva de la historia
When you say the algorithm quadrupled misogynistic content in five days, what does that actually look like on someone's screen?
It means if you're a young man searching for help with loneliness or mental health, the platform starts feeding you content that dehumanizes women—jokes about violence, claims about women's bodies, the whole spectrum of it. And it does this because engagement is the metric that matters. Outrage keeps people scrolling.
But TikTok says they remove 93 percent of content that violates their rules. Doesn't that suggest the system is working?
It suggests they're catching some of it. But "some" isn't the same as "most," and it certainly isn't the same as preventing the radicalization pipeline. A young man looking for community around loneliness finds it—but it's a community built on hating women. By the time he's seen enough of it, the hatred feels normal.
You mention 154,000 additional girls reporting forced sex. That's a staggering number. How confident are you that social media is actually causing that increase?
I'm not saying social media is the only cause. But research is clear: exposure to violence in media correlates with committing violence. And we've created an environment where violence against women is not just present—it's celebrated, algorithmically amplified, and normalized. That has to be part of the equation.
What strikes me is the policy angle—the childcare funding, the nutrition programs. How does online misogyny connect to Congress letting billions expire?
It's about what becomes politically possible. If the culture has shifted to accept and even glorify misogyny, then depriving women of resources stops feeling like cruelty and starts feeling like acceptable policy. The online world shapes what we think is normal. That shapes what we vote for.
You mention "calling people in" rather than fighting trolls online. That feels like a lot of emotional labor to ask of women.
It is. But the alternative—engaging publicly with trolls—just feeds the algorithm more engagement, which spreads the content further. The real work happens offline, in conversations where someone might actually listen. It's not fair that women have to do this work. But it's more effective than the alternative.