By DAVID KLEPPER

PROVIDENCE, R.I. (AP) — It should have been a time of celebration: Brittan Heller would soon graduate from college and head to one of the nation’s top law programs.

But when a classmate with unrequited feelings for Heller wasn’t admitted to that same school, he turned his rage on her. He wrote a manifesto titled “A Stupid B—h to Attend Yale Law School” and posted it on a site popular with anonymous trolls. The man urged them to do their worst.

Soon strangers were making derogatory, sexualized comments and posting her pictures online. They made threats. Posted her personal information. At one point, FBI agents escorted Heller to class for her protection.

“People say, ‘Oh, just log off. Don’t read it. Turn off the computer,’” said Heller, who turned her personal experience from 15 years ago into a legal specialty as a leading expert on online harassment. “This the 21st century, and people have a right to use the internet for work, for pleasure or to express themselves. Telling people not to read the comments is no longer enough. We don’t talk enough about this problem, and we need to.”

Online harassment has become such a familiar part of the internet that it can be hard to imagine the web without it. From teen cyberbullying to authoritarian governments silencing dissent, online toxicity is a fact of life for everyone, with women, teens and religious and racial minorities the most likely to be targeted.

And there is evidence the problem is getting worse.

In 2014, 15% of Americans said they had faced severe or significant online abuse, defined as stalking, physical threats, sustained harassment or sexual harassment. In 2021 the number was 25%, according to studies by the Pew Charitable Trusts.

Health care workers, journalists, teachers, police and government workers have all reported increases in online harassment in recent years, as the pandemic and political polarization led many people to release their anger and fear online.

Nearly three in four female journalists reported receiving threats or other forms of online harassment, according to a survey by UNESCO and the International Center for Journalists that polled more than 700 journalists in more than 100 countries. One in five of those said the harassment escalated to offline abuse or even assault.

The growth of the internet has also expanded the ways that people can be targeted beyond merely email to social media posts, direct messages, texts and streaming video. And with the rise of smartphones and cheap, ubiquitous internet, harassment can now be a 24-7 problem for victims.

“We’ve made so many strides — there’s more awareness now — but it’s easy to get frustrated and to feel like we’ve gotten nowhere,” said Tina Meier, who started a foundation to teach kids and parents about online harassment after her daughter’s suicide in 2006.

Thirteen-year-old Megan Meier had been bullied by someone she met online who she thought was a teenage boy named Josh. The two had flirted until the person suddenly turned against Megan. “Everybody hates you,” “Josh” wrote. “The world would be a better place without you.”

Police later determined that “Josh” was actually an adult woman, the mother of one of Megan’s classmates.

While polls show all types of people are susceptible to online harassment, extensive research has shown that women and people of color are far more likely to be targeted. That’s also true for people with disabilities, people who belong to religious minorities and members of the LGBTQ community.

Women are more likely than men to say online harassment is a serious problem, Pew found. They’re also more likely to report being the victims of online sexual harassment and more serious abuse such as threats of physical harm.

The difference is so great that many men may not understand the severity of the demeaning language, sexualized insults and unwanted attention that women frequently face online. A coordinated harassment campaign against female video game designers that began in 2014, known as Gamergate, became so pervasive — including threats of rape, torture and murder — that some women hired security or went into hiding.

Online harassment has also been used globally to attack journalists, dissidents and others in the public arena.

Political consultant Maria Cardona began receiving nasty emails and direct messages once she began presenting her opinions on national news shows. She’s noticed that many of her critics seem focused on the idea that an outspoken Latina woman could be considered an authority on politics.

One typical message read: “I hope you get raped and have your throat slit.”

“They want to shut us up, they want to scare us, they want to intimidate us,” said Cardona, who now keeps her office locked after someone showed up to accost her in person.

Anonymity can make it easier to be cruel without fear of offline repercussions. It’s a phenomenon called the online disinhibition effect, and it’s one reason why trolls feel comfortable saying things they would never say to someone in person.

As part of a 2009 settlement of Heller’s lawsuit against her harassers, she asked to meet them face to face. One was a 17-year-old boy who had posted that he’d like to gouge Heller’s eyes out and have sex with her corpse.

“They all essentially said the same thing: that they didn’t realize their actions were impacting a person in that way, that they didn’t realize there was a person on the other side of the screen,” Heller recalled. “And they all said, ‘I am so sorry.’”

California enacted the nation’s first law against cyberbullying in 1999, and most states have since followed suit. Enforcement can be difficult, however, as the lines between harassment and free speech can be blurry. Police and prosecutors often lack sufficient training or resources.

Tech companies say they are getting better at identifying and stopping harassment. For example, Instagram, which is owned by Facebook parent company Meta, made several changes designed to reduce harassment, including putting warning labels on potentially abusive language and making it easier to block or report harassers.

Yet those moves haven’t been enough. Internal Facebook documents leaked by former employee Frances Haugen show that executives are aware of the potential for their products to be used to harass people. One internal study cited 13.5% of teen girls saying Instagram exacerbates suicidal thoughts and 17% saying it worsens eating disorders.

“Online harassment is a problem for everybody, but I think it’s especially problematic for kids,” said Natalie Bazarova, a professor at Cornell University who studies social media.

She said a multifaceted approach is required to address the problem: legislation to require minimum safeguards from tech companies, technical innovations and extensive educational efforts such as simulations that teach teens to spot cyberbullying and use social media safely.

Technical solutions include automated systems that flag posts for signs of harassing language — all-capital letters, repetitive phrases, certain key words — or instituting a short delay before users can respond to posts, giving them a chance to cool off.

Now in its “awkward adolescence,” the internet is not the first invention to change how humans communicate, Heller said.

“People said similar things about the telegraph, the telephone and the television — that they were somehow going to ruin society,” she said. “They were all regulated about 25 years into their life cycle. Those regulations didn’t kill the telephone, the television or the radio.”