The Complete Guide to Pig Butchering Scams

The text arrives on a Tuesday afternoon. "Hey, is this Michael?" Or: "Sorry, I think I got the wrong number, but are you from Austin? You look familiar." Or: "It's been too long, let's catch up." Each message seems random. None of them is.

Pig butchering is a coordinated, industrial-scale fraud operation with roots in forced-labor compounds in Cambodia, Myanmar, Laos, and the Philippines. Victims in the United States alone lost an estimated $5.5 billion to it in 2024, across roughly 200,000 cases. The average victim lost $27,500. One victim in Brooklyn, the retired insurance agent Ron Williams, whose case NBC News covered in 2026, lost $1.6 million after a text from a fictional woman named "Jenny." A separate DOJ forfeiture action in 2025 seized 127,271 bitcoin worth $15 billion from a single Cambodian conglomerate, Prince Group, whose chairman was charged with running the compounds.

If you understand how the scam works, you can stop it before it starts. This is the complete guide.

What is pig butchering?

Pig butchering, a direct translation of the Chinese term shā zhū pán (杀猪盘), is a long-duration romance-and-investment scam that unfolds in stages. The name comes from the agricultural practice of fattening an animal for weeks before slaughter. The victim is the "pig." The "fattening" is the weeks or months of emotional investment the scammer builds before asking for money. The "slaughter" is the final transfer, often of a victim's entire savings, to a fake investment platform.

The distinguishing feature of pig butchering, compared to traditional romance scams or investment scams, is its duration and its commercial polish. A pig-butchering scam is not a single message. It is a relationship. And by the time the financial ask arrives, the victim has usually been texting the scammer daily for weeks or months. The emotional investment makes it nearly impossible to walk away at the moment skepticism should kick in.

Where does the term come from?

Chinese-speaking criminal networks in Southeast Asia coined the term internally. The metaphor, one that scam-compound workers themselves used in training materials, describes the business model with disturbing precision. Western journalists picked up the term in coverage beginning around 2021, and it is now standard in FBI, DOJ, and international law-enforcement reporting.

How the scam unfolds, week by week

The approach (day 0 to day 3)

The opening contact almost always arrives unsolicited. The most common openers: a "wrong number" text, an unexpected message on WhatsApp or Telegram, a DM on LinkedIn or Instagram, a match on a dating app that progresses unusually fast. The scammer assumes the target is a real person in their twenties to seventies with disposable income. Targeting data is purchased or scraped.

If the target engages, even to say "wrong number," the scammer pivots. "Oh, so sorry, you seem nice though. I'm [name], I'm [age], I'm [profession] in [city]. What do you do?" Something that would be an imposition in person works online because it takes no physical energy and feels safe.

The build (week 1 to week 4)

The scammer and the target exchange messages daily. Morning texts. Food photos. Inside jokes. Small vulnerabilities. The scammer's persona, often a successful, attractive, ambitious young woman in the case of an older male target, or a steady, kind, slightly-older-than-expected man in the case of a single female target, reveals itself gradually. Real details appear about family, schooling, religion. Some of these details will later prove to be stolen from a real person; some are entirely fabricated. Voice notes may appear. Sometimes a brief, carefully-arranged video call.

By the end of the first month, most targets feel they have a real relationship. Some have said "I love you."

This is also the stage where AI tools have most dramatically scaled the industry. The scammer is usually not communicating with one target. They are communicating with dozens. Language models auto-generate conversation drafts tuned to each target's profile. Voice clones produce a consistent "voice" across voice messages even when the operator changes shifts. Real-time deepfake video makes the rare agreed video call look plausible. The supervisor at the scam compound can handle volume that was impossible in 2020. For more on the synthetic-persona side of this industry, read how AI personas scam you on social media.

The setup (week 5 to week 8)

The scammer "mentions" an investment. A cryptocurrency arbitrage opportunity her uncle introduced her to. A gold-trading platform his sister uses. An AI-powered trading bot a mentor recommended. Returns of 20%, 40%, sometimes 200%. The scammer is cautious at first, don't invest anything you'd miss, which reads as trustworthy advice. A small test transaction is suggested. The target sends $500 or $1,000 to a "trading platform" that looks professional and shows an account dashboard updating in real time.

Over a week or two, the dashboard shows the balance grow. The scammer expresses delight. She or he urges the target not to leave money on the table but to reinvest the gains. The target tries a small withdrawal. It works. The funds arrive in the target's bank account. Trust, which was emotional, now has a financial proof point. The scammer has structured the scam so that the first small withdrawal is real. That is the hook.

The slaughter (week 9 to week 16)

Now the scammer asks the target to go larger. Maybe a home-equity line. Maybe retirement savings. Maybe a loan from family. The scammer's emotional presence intensifies, dreams about the future together, plans to visit. The dashboard shows the balance climb toward life-changing amounts. $400,000. $1 million. $4 million, in Ron Williams's case.

Then the target tries to withdraw the big number. The platform tells them they owe tax, or a withdrawal fee, or an international transfer verification charge. The scammer is sympathetic and urgent, pay the tax so we can release the gains. The target sends more. The platform tells them another fee is required. The scammer pushes again. The target sends more.

At some point, the target refuses or runs out of money. The scammer disappears. The platform goes offline. The phone number stops responding. There was never any investment, never any balance, never any Jenny.

The aftermath

By the time the victim understands what happened, the money is almost always unrecoverable. Crypto transactions do not reverse. The compound structure and multi-hop laundering, sometimes through regulated exchanges, sometimes through overseas shell companies, sometimes through bitcoin mixers, makes tracing funds technically possible but practically useless for most victims. A small number of large cases result in partial recovery. Most victims see none of the money again.

Where pig butchering actually comes from

The modern pig-butchering industry is not a cottage industry of individual con artists. It is a set of industrial operations.

The most-documented operator is Prince Holding Group, a Cambodian conglomerate founded by Chen Zhi. In October 2025, the U.S. Department of Justice unsealed an indictment charging Chen with wire fraud conspiracy and money laundering conspiracy for directing Prince Group's operation of "forced-labor scam compounds across Cambodia." The compounds, according to the indictment, were "vast dormitories surrounded by high walls and barbed wire" that "functioned as violent forced labor camps." Workers were trafficked in, often under false job offers, then prevented from leaving. Two specific compound sites, per court filings, housed 1,250 mobile phones controlling 76,000 social-media accounts. The DOJ's simultaneous civil forfeiture of 127,271 bitcoin, worth $15 billion, was the largest forfeiture action in its history.

Chen Zhi was arrested in Cambodia and extradited to China in January 2026, triggering what reporters described as a "mass exodus" of workers from several compound sites. But the industry spans jurisdictions. The United Nations Office on Drugs and Crime has estimated that more than 100,000 people are held in scam compounds across Southeast Asia, with operations documented in Myanmar's Shan State, Laos's Golden Triangle Special Economic Zone, and parts of the Philippines. Arresting a single chairman does not dismantle the infrastructure.

For the full story of the Chen Zhi indictment, see our Case File: Chen Zhi and Prince Group: The $15B Pig-Butchering Empire.

The role of AI in modern pig butchering

Every stage of the scam now runs on AI.

Targeting uses machine-learning classifiers to segment victims by age, language, likely disposable income, and emotional-vulnerability signals scraped from social media.

Conversation runs on language models that produce tailored messages at scale. A single compound worker can maintain plausible ongoing relationships with dozens of targets simultaneously, with each target receiving personalized content.

Voice uses commercial voice-cloning tools to produce consistent audio messages regardless of which compound worker is on shift.

Video uses real-time deepfake overlays so a rare video call appears to show the persona the target has been texting, the same face, the same voice, the same mannerisms, regardless of who is physically on camera.

Platforms, the fake trading dashboards, the counterfeit "app" the target is instructed to install, are generated from templates that can be spun up in minutes and torn down the moment the victim stops paying.

The practical consequence for a target is that every traditional "proof of authenticity" has been neutralized. A voice that sounds like Jenny is not proof it is Jenny. A video that shows Jenny is not proof it is Jenny. An official-looking investment platform is not proof the platform exists. The only reliable verification is external, independent, and offline.

How to recognize a pig-butchering opener

Most pig-butchering scams begin with one of a small set of opening messages. If you receive any of these, the safest assumption is that you are a target.

• "Hey, is this Michael?" or "Sorry, wrong number, but you seem nice." Block, delete, and do not reply.
• "Long time no talk!" or "It's been too long!" from a number you do not recognize. Block, delete, and do not reply.
• "I had dinner with your cousin last week, she told me about you." Block, delete, and do not reply. Even if a cousin exists, a real friend-of-a-friend sends a message through your cousin, not an unsolicited text.
• "I saw your profile, we matched and I wanted to reach out." If you have no memory of the match and no matching context, block.
• An unusually attractive match on a dating app who wants to move the conversation off-platform within 48 hours. Off-platform means WhatsApp, Telegram, WeChat. This is the single most consistent tell in dating-app pig butchering.

The correct response to every one of these is no response. Blocking and deleting costs you nothing. A reply, even "wrong number," tells the targeting system your phone number is live. For the broader pattern, see how to tell if a text message is a scam.

What to do if you are targeted

If the conversation has just started: block and delete. Do not engage. Do not try to confront or "waste their time." Scam compounds have efficient pipelines for re-targeting users who engaged; confrontation only confirms you are a live target.

If you have sent any money: contact your bank and credit-card company immediately. For cryptocurrency, contact the exchange you sent from, funds that have not yet been laundered can sometimes be frozen if reported within hours. File a report at IC3 and reportfraud.ftc.gov.

If you have sent significant money: engage a lawyer experienced in financial fraud, and report to the FBI's IC3. In cases involving large crypto transfers, blockchain-tracing firms sometimes can identify recipient wallets that are then subject to forfeiture action.

If a family member has been targeted: be patient. Victims often refuse to accept what has happened until they see concrete proof, in Ron Williams's case, his son had to generate a new AI "Jenny" from scratch to demonstrate how trivial the technology is. Shame is the scammer's ally. Your job is to get the victim out of the scam and into reporting, not to litigate how it happened. For elder-specific protection, see how to protect elderly parents from scams.

How to protect yourself from the start

1. Never respond to cold text messages from unknown numbers. The cost of responding is much higher than the cost of ignoring.
2. Never invest in a platform recommended by someone you met online. Full stop. No exceptions. If the opportunity only exists inside the conversation, it is the scam.
3. Treat every "gains" dashboard as fake until you successfully withdraw to your own bank. The only real test is a real withdrawal. Scam platforms design the first small withdrawal to succeed; do not let that be your proof of trust.
4. Treat every "you need to pay taxes first" message as confirmation the platform is a scam. This is the most consistent closing move in the industry.
5. Agree a family code word. For anyone who might be contacted with a fake emergency, a code word cuts through voice cloning and deepfake video. A voice that sounds right but does not know the word is not the person.
6. Read our companion guide on signs of a romance scam.
7. When anything feels off, scan it. Paste suspicious texts, profiles, platforms, or promotional content into AuthentiLens. We flag pig-butchering language patterns, AI-generated media, and social-engineering signals in seconds.

Resources for victims

• FBI Internet Crime Complaint Center (IC3), file any online financial crime report here
• Federal Trade Commission, report any consumer fraud
• Chainalysis Crypto Crime Reports, industry research on crypto-enabled fraud flows
• Global Anti-Scam Organization (GASO), victim-support nonprofit focused on pig butchering
• AARP Scam Resources, elder-focused fraud prevention