The dating scene in Los Angeles is tough. Doubly so if you’re not fresh out of college. That’s what Evelyn found when a relationship of two decades came to an abrupt end last year, casting her back into the dating pool at 50.

“All the beautiful twentysomethings are here trying to make it in the entertainment business. And men of all ages are trying to date them,” says Evelyn, who asked to appear under a pseudonym. “Even among younger folks, I’m sure it’s hard, but when you’re my age it’s definitely a tougher market.”

Evelyn’s family emigrated from Korea when she was young and lived in various parts of the Los Angeles area. In the early 2000s, she moved with her partner to a leafy suburb in the Hollywood Hills, “an enclave of hippies and music lovers.” Evelyn was just about the only Asian there, but she found the area tranquil.

When the relationship fell apart, Evelyn was forced out of her home and out of early retirement too. It was a terrible period, she says, but returning to dating was a way of “not taking it lying down.” Like many thousands of others, she didn’t feel like going to bars and clubs and turned instead to dating app Hinge. She was nervous—how do you reject someone’s advances politely, she wondered, or tell them you don’t want a second date—but she was eager to try.

One day in late April, she matched with Bruce Zhao, a Chinese man of a similar age who dressed well and had kind eyes. He lived in Sunnyvale, a few hours’ drive up the coast, but he had a second home in Los Angeles. He was sweet, attentive, and not at all aggressive, which was what Evelyn liked most.

The conversation was occasionally stiff—“Good morning, a new month has started, I wish you all the best,” Bruce wrote in one message. “Whether we end up as lovers or friends, I believe we will have more interactions in the future,” he said in another. He had a habit of mansplaining too. But Evelyn found herself wanting to impress. “I felt so rejected from my previous relationship. I just wanted to feel accepted and liked,” she says.

Over the next five weeks, the pair swapped messages over WhatsApp almost every day. Smiley emojis turned to winky ones, then to kissy ones. Though they never met in person and spoke only once over the phone, the terms “boyfriend” and “girlfriend” entered their vocabulary. Just like in the ads, they deleted Hinge together.

Before long, the conversation turned to work. Bruce told her he was a partner at a venture capital firm but ran a crypto trading studio on the side. If Evelyn liked, he could introduce her to investing—starting small, with $4,000? Evelyn said that made her nervous, so they settled on half the amount.

To pull off the trading strategy he had planned, Evelyn would need to use a special crypto exchange, called CEG. Bruce walked her through the process: moving cash from the bank into Coinbase, a regular exchange, then converting that cash into crypto and sending it to CEG. Evelyn was skeptical at first, but to her eye, the website looked legitimate. The layout was neat and the interface well-designed, there was a customer support function, and the feed updated in real time as the value of cryptocurrencies changed.

Bruce instructed her to perform a sequence of trades, with precise timing, betting on either an increase or decrease in the price of bitcoin. He had studied “the implementation of the international market trends,” he said, to tabulate the “precise time fluctuations of the rise and fall” of the coin. Evelyn couldn’t follow what was going on, and the whole thing made her heart race, but within half an hour, she was up by 20 percent: “Dude. I can see how this can be addicting,” she said to Bruce.

To minimize risk and maximize profit, Bruce said, it was better to play with larger sums: at least $200,000. Evelyn didn’t have that kind of cash on hand, so Bruce suggested that she she tap her individual retirement account (IRA). At first, she recoiled—“You’re asking me to undo 40-plus years of societal financial teachings in 30 minutes,” she said—but later she worried her refusal might jeopardize the relationship. “Will we still continue to get to know each other, even if the $200K doesn’t happen?” she asked the next day.

Bruce and Evelyn traded together on CEG on four more occasions. She drained her savings, committing $300,000 in total and eating a 10 percent penalty for the early IRA withdrawal, but the website said her investment had grown to over half a million dollars. Bruce even helped her cash out a small sum to go shopping in celebration. Between trades, they continued to chat over WhatsApp, about Mexican food, the Lakers, TikTok, Elon Musk, their favorite movies, and other artifacts of normal life.

In all likelihood, Bruce Zhao does not exist. His persona, certainly, is a calculated fiction, devised to prey on the vulnerabilities of people in search of intimacy and dependent on digital services to find it. The “trading sessions” were an elaborate pantomime.

Romance scams predate the internet by centuries. In the 16th century, wealthy people were scammed by letter. But since the pandemic, which drove social isolation and forced more interactions online, fraud has reached epidemic levels. Scams have also evolved with the age of digital communication into a far more potent threat. Without the physical cues available in person, people have less information to help identify a threat, according to Gareth Norris, a lecturer in psychology at Aberystwyth University who published a paper in 2019 on the psychology of internet fraud. Behavior that might normally raise suspicion—sweating, say, or fidgeting—is concealed behind the screen.

The architects of these scams have made a profession of hijacking human psychology, says Norris, and are able to manipulate the way people make decisions for their own gain. For example, a natural tendency toward confirmation bias, focusing on cues that affirm an already formed perception, can be abused to squeeze more money out of a victim. “Human beings don’t tend to like ambiguity; they don’t like to hold two competing views in mind. When we’ve got a suspicion something isn’t right, it creates tension in our brains,” says Norris. “But confirmation bias helps avoid ambiguity by only focusing on information that supports our view and ignoring things that don’t.”

The likelihood that someone will fall for a scam has “no connection with intelligence,” says David Modic, a professor of computer and information science at the University of Ljubljana. Rather, it’s about the ability of the scammer to take advantage of personal context—like a recent breakup—and social engineering techniques to “erode someone’s self-control” and blind them to warning signs.

That’s what Evelyn found: “There were red flags,” she says, “but no smoking gun that outweighed my desire to stay in touch with [Bruce].”

Evelyn caught the first glimpses of reality when she asked Bruce to help her withdraw a substantial sum. He acted offended and implied that she was ungrateful: “Why are you always in doubt?” he asked. Evelyn tried to perform a withdrawal without his help but received an error on the CEG website. The online customer support was useless.

Beginning to suspect something, Evelyn did a reverse Google search of an image of Bruce’s office. It had been posted by someone else on Twitter, two years earlier. The owner of the account was using it to promote a separate trading operation, AstroFX, another alleged scam. A deep uneasiness began to swell in the pit of her stomach.

The last vestige of hope evaporated when Bruce told her that to withdraw her profits, she would have to deposit an additional 15 percent of the value of her account, supposedly to verify her identity. In a fit of absent-mindedness, he had forgotten all about the requirement. When she said she didn’t have the funds, he told her to borrow from friends. “What are you going to do, dear?” he wrote, pressing her.

For Evelyn, looking back on her interactions with Bruce is like watching a slow-motion car crash. “There was no way this guy found a way to print money, and yet I kept going,” she says. “I have so much PTSD driving by the local bank where I made the wire transfers. I walked to the bank on my own two feet. It kills me to drive by.”

Evelyn’s bank, Wells Fargo, said it couldn’t reimburse her because she had voluntarily triggered the wire transfers. The same was true of Coinbase, the exchange she used to trade dollars for crypto, because she had willingly sent her coins elsewhere. She met with the Los Angeles Police Department in the hope that it might lead to an investigation and even the return of her funds. But a detective told her she’d given them nothing to go on because the name, location, and other personal details provided by Bruce were probably fake. The investigation ground to a halt before it had even begun.

The LAPD did not respond to a request for comment. Coinbase declined to comment. Jim Seitz, a spokesperson for Wells Fargo, says the bank is “actively working to raise awareness of common scams.” And Jarryd Boyd, director of brand communications at Hinge, says the company has “sophisticated machine learning technology in place and trained content moderators that are constantly patrolling for fraud” and that it will continue to invest in new technologies to help keep users safe.

This particular kind of investment scam, where the scammer forms a relationship with their victim over an extended period, usually under romantic pretenses, is known as pig butchering. It’s alarmingly common.

The modus operandi, says Adrian Cheek, an independent security researcher based in Canada, is almost always the same: A cybercriminal organization develops a template for a fake crypto trading site and, in exchange for a kickback, hands it to other criminal affiliates who perform the scam. Data gathered by Cheek suggests that at least 439 distinct web addresses have been used to host the CEG exchange. “It’s absolutely textbook,” he says.

The people interacting with the victim—the Bruces—are often themselves victims of a different kind, trafficked into Thailand, Myanmar, or Nigeria and taken into forced labor. Lured with false promises of employment, they are “forced to sit in call centers that are essentially prisons and contact people endlessly,” says Cheek. At any one time, they might be speaking to hundreds of different targets, following a set script. “It’s a continual cycle of victims,” says Cheek. “They might be starting today on a new victim but at the same time be about to close out another. It’s work, work, work.”

The double victimization can make it even more difficult for those whose money has been stolen to process what has happened to them because it robs them of a focal point for their rage. A faceless cybercriminal organization is far harder to hate than someone with a name, face, and dating profile. Evelyn says knowing that she might be up against a large, organized entity makes retaliation feel futile.

The likelihood that a victim will get their money back is also slim; the only hope lies in tracing the movement of cryptocurrency. If law enforcement is able to identify which exchange the criminals use to cash out into regular money, there is a chance they might identify the individual or group responsible for the scam.

An analysis of the movements of the crypto stolen from Evelyn conducted by Chainalysis, a firm whose tools are used by law enforcement to inform investigations, reveals the lengths scammers will go to cover their tracks. After Evelyn handed over the crypto, it was divided into different wallets (which can be thought of like bank accounts) and eventually cashed out via a selection of exchanges based in multiple countries. But along the way, it traveled through up to five different wallets, where it was blended with takings from other victims and converted on multiple occasions into different crypto tokens. Each of the hops is designed to further obfuscate the origin of the funds and limit the likelihood an exchange might identify wrongdoing and freeze the assets. “This indicates a high level of organization,” says Phil Larratt, director of cybersecurity investigations at Chainalysis. “But it’s become more common, in the last 12 months, for scammers to use this kind of methodology.”

As a matter of policy, Chainalysis declined to name the exchanges used by the CEG scammers. But in any case, says Larratt, criminals frequently use money mules or stolen identity documentation to open the exchange accounts used to cash out, which means identifying the owner of an account is not the same as identifying the scammer.

The sheer volume of scams like this one, coupled with the lack of resources and technical expertise among law enforcement, means investigations are rare. “These scams are extremely hard to investigate, prosecute, and disrupt,” says Nicola Staub, a former cybercrime prosecutor and now CEO of security startup Cybera. “It’s all about speed.” The faster a victim can have somebody find the exchange, the better their chances of recovering funds. “But most that turn to law enforcement don’t get help with tracing, or don’t get it fast enough,” Staub says.

Even if local police employ specialists or external partners to trace the movement of crypto, says Staub, the “extremely international” nature of scam operations means it can take months or years to complete the paperwork and processes necessary to trade information between jurisdictions.

The estimates vary, but all told, only a minute percentage of victims ever see their money again. Others are revictimized by recovery scammers who ask for upfront payment in return for help retrieving lost funds.

It has been difficult for Evelyn to return to her regular life, as she is still hurting from the breakup and now has the loss of her savings hanging over her. In her darkest moments, she wondered whether it might be simpler to kill herself. “That was the lowest I felt, when I really embraced the finality and reality of the situation,” she says.

Not wanting to be branded an idiot, Evelyn has barely told anyone what happened. But with the help of a therapist, she is probing for the root of her vulnerability. The “overriding feeling,” she says, when she began to speak to Bruce, was a desire to “play nice.” But in therapy, she hopes to learn to say no.

Evelyn plans to return to Hinge too—eventually: “I’ve learned my default mode is to put one foot in front of the other. I haven’t let this completely defeat me.”