Romance scams: How they work, how they win and what we do about it

2026-02-18 · Source: IBM Technology · Field: Technology & Digital — Cybersecurity & Data Privacy, Artificial Intelligence & Machine Learning · Depth: Intermediate, extended

Summary

Romance scams, a pervasive form of social engineering, cost Americans $672 million in 2024 and represent a multi-billion dollar global industry. These scams are typically orchestrated by complex organized crime groups, not lone individuals, often involving human trafficking where victims are forced to text targets for 17 hours daily. Tactics include "wrong number" texts, catfishing on dating apps, and "pig butchering" long cons that cultivate emotional relationships before soliciting funds, frequently for cryptocurrency investments. AI tools, including natural language generation and deepfakes, enhance scammers' ability to create believable personas and automate victim research by exploiting public records like divorce filings or layoffs. High-profile individuals, including a CEO who embezzled $47.1 million for a crypto scam, and retirees losing life savings, are equally vulnerable, challenging the misconception that only "gullible" or older individuals are targeted. The emotional manipulation makes these scams particularly difficult to detect and interrupt, often leading to significant financial and personal ruin.

Key takeaway

For security leaders and financial executives, recognizing romance scams as a sophisticated organizational threat, not just a personal one, is crucial. Implement robust employee awareness programs that specifically address the emotional and financial tactics of these scams, emphasizing that anyone can be a target. Additionally, advocate for and deploy advanced fraud detection controls within financial institutions, particularly for cryptocurrency transactions, to disrupt large money movements and mitigate harm, as technological solutions can help where emotional firewalls cannot.

Key insights

Romance scams are a multi-billion dollar organized crime industry leveraging emotional manipulation and AI to defraud diverse victims.

Principles

• Emotional manipulation bypasses intellectual defenses.
• AI enhances scam sophistication and targeting.
• Stigma prevents victims from seeking help.

Method

Scammers cultivate long-term virtual relationships, often via "wrong number" texts or dating apps, then introduce fraudulent investment opportunities, typically in untraceable cryptocurrency, after building trust.

In practice

• Do not respond to unsolicited texts or calls.
• Be skeptical of online investment opportunities.
• Implement fraud detection for anomalous transactions.

Topics

• Romance Scams
• Social Engineering
• Organized Cybercrime
• AI in Scams
• Cryptocurrency Fraud

Best for: CTO, Executive, Security Engineer, IT Professional, General Interest

Related on AIssential

• Fraud Has Moved Outside the Bank. Protection Should Too — Fraud has moved beyond bank perimeters, necessitating device-level protection and cross-industry intelligence sharing to combat AI-enabled social engineering.
• AI Is Turbocharging the Spamosphere, Amping Up Prolific Text-Message Scams — AI, particularly Google's Gemini model, is being exploited to create sophisticated fake websites, turbocharging text-message scams and financial losses.
• Southeast Asia Scam Compounds Turn AI Into a Cybersecurity Threat — AI and automation are scaling Southeast Asian scam compounds, making fraud more sophisticated and harder to detect.
• AI-powered scams drove surge in fraud attempts during 2025 — AI-enhanced scams are driving a significant surge in fraud attempts and consumer financial losses.
• Ocean raises $20 million Series A to tackle AI-driven email attacks - CTech — AI-driven email attacks necessitate advanced AI defenses capable of understanding intent and context, surpassing legacy pattern-based security.
• 244,000 AI Developers Downloaded What They Thought Was OpenAI and It Stole Their Crypto Wallets and… — A sophisticated supply chain attack exploited trust in AI platforms by mimicking a legitimate OpenAI tool to steal sensitive user data.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by IBM Technology.