Generative AI has transformed traditional family scams into sophisticated, scalable fraud networks, posing a new era of danger for all, especially vulnerable older adults, as law enforcement warns of a rise in convincing deception tactics.
A family scam that once relied on patience and a few persuasive emails now has a far more dangerous ally. Generative artificial intelligence can draft polished messages, translate in real time, produce false documents and photographs, and even mimic a familiar voice, allowing criminals to sustain believable contact with many victims at once. The result is not a new kind of con, but a more efficient version of an old one.
INTERPOL has described this shift as the “industrialisation of fraud”. In its 2026 Global Financial Fraud Threat Assessment, the organisation said AI-enabled fraud is 4.5 times more profitable than traditional methods because it lets criminals contact more people while sounding more convincing. It also warned that so-called agentic AI, which can plan and carry out sequences of tasks with limited human direction, could automate much of a fraud campaign.
The FBI says the threat is already showing up in its data. In its 2025 Internet Crime Report, the bureau recorded 22,364 complaints involving AI-related information and more than $893 million in adjusted losses. It said AI is helping criminals create synthetic profiles and personalised conversations at scale, while high-quality fake content is becoming easier to generate and harder to spot. Older adults were hit particularly hard: people aged 60 and over filed just over 3,100 of those complaints, but reported losses of more than $352 million, or well over a third of the total.
Recent disclosures from OpenAI illustrate how the technology is being used. In February, the company described “Operation Date Bait”, a Cambodia-linked romance and task scam that used AI to generate advertisements aimed at men in Indonesia. An AI chatbot played the role of a flirtatious receptionist, while human operators used machine-generated and translated messages to draw victims into escalating payments. In July, OpenAI said another Cambodia-based network used ChatGPT in investment, romance, gambling and law-enforcement impersonation schemes, including fake passports, legal notices and investment confirmations.
The risk is not confined to tricking people into sending money. Google’s Threat Intelligence Group said in May that it had seen what it described as the first known case of a criminal using a zero-day exploit, meaning a flaw that had not yet been publicly identified, developed with AI. The attack targeted a widely used website-management tool and was designed to get past two-factor authentication, the extra code used after a password. Google said its engineers stopped the plot before it could be deployed at scale.
For families, the lesson is that both machines and people are now attack surfaces. The FBI advises Americans to pause before reacting to urgent demands for money or information, and to verify requests through a separate channel. The Federal Trade Commission has also warned that voices can now be cloned convincingly enough to fool relatives. A private verification phrase, multi-factor authentication, and a refusal to act on pressure, secrecy, gift cards or cryptocurrency demands are no longer optional precautions; they are basic self-defence in an era when deception can be industrialised.
Disclaimer: This article is intended to inform and educate, not to recommend or endorse any financial product, investment or strategy. Please consider your own financial circumstances and seek professional advice where appropriate before making financial decisions.
