Dangerous Scams Spreading Through Technology Right Now

AI Voice Cloning and Deepfake Technology in Scams

Among the most alarming developments in online fraud is the industrialization of voice cloning and deepfake video technology. Tools that once required expensive hardware and specialized expertise can now generate a convincing replica of any person’s voice from as little as three seconds of audio, according to current research cited by cybersecurity firm CybelAngel. In one documented case, engineering firm Arup lost $25.6 million after employees were fooled by a deepfake video call in which a fraudster impersonated a company executive. The FBI’s 2025 Internet Crime Report logged more than 22,000 AI-related fraud complaints with losses exceeding $893 million in that category alone — though Congressional researchers estimate fewer than five percent of voice clone victims report their losses, suggesting actual damages are considerably higher.

Deepfake-related losses in the United States reached $1.1 billion in 2025, tripling from $360 million in 2024, according to data compiled by Keepnet Labs. Deepfake-enabled voice phishing attacks — known as vishing — surged by over 1,600 percent in just the first quarter of 2025 compared to the fourth quarter of 2024. The FBI has issued public warnings about campaigns in which AI-cloned voices are used to impersonate senior U.S. officials, combining text messages and synthetic audio to steal sensitive information. In early 2026, cybersecurity researchers at Google documented attacks in which the Gemini AI tool was used to conduct reconnaissance and generate deepfake personas for use in fraudulent video calls, including one case linked to a North Korean state-sponsored group that used an AI-generated version of a prominent CEO to compromise a victim’s computer.

Personal family impersonation — sometimes called “grandparent scams” or “virtual kidnapping” — has taken on a disturbing new dimension with AI voice synthesis. Fraudsters harvest audio samples from social media videos to clone the voice of a family member, then call relatives claiming an emergency and demanding immediate wire transfers or gift card payments. The FBI documented this growing trend in formal advisories in 2025 and 2026, noting that older Americans are disproportionately targeted.

Pig Butchering: The Cryptocurrency Romance Scam Destroying Savings

The FBI reported that Americans lost $9.3 billion to cryptocurrency-related crimes in 2024 — a 66 percent jump from the year before — with investment fraud accounting for $5.8 billion of that total. A significant and growing share of those losses is attributable to a scheme known as “pig butchering,” a term that refers to the process of “fattening up” victims emotionally before draining their finances. The scam typically begins with an unsolicited message on social media, a dating application, or a messaging platform like Telegram. The fraudster, often operating from organized criminal compounds in Southeast Asia, builds a genuine-seeming friendship or romantic relationship over weeks or months before introducing a supposedly lucrative cryptocurrency investment opportunity.

Once a victim transfers funds to a fraudulent trading platform, they often see convincing fake returns displayed on the platform’s interface — a deliberate tactic designed to encourage larger deposits. When the victim eventually attempts to withdraw their money, the platform fabricates tax obligations or “fee” requirements that require further payments, and ultimately disappears entirely. TRM Labs estimated that pig butchering scams siphoned over $4.4 billion in a single year, while noting that only about 15 percent of victims ever report their losses. The U.S. Department of Justice has increasingly moved against these operations: a coordinated international takedown announced in May 2026 led to at least 276 arrests connected to scam centers operating under names including “Ko Thet Company,” “Sanduo Group,” and “Giant Company,” according to the DOJ’s official press release. The FBI also seized $8.2 million in cryptocurrency directly linked to pig butchering schemes on dating applications.

Generative AI has accelerated the pig butchering model considerably. Fraudsters now use AI chatbot tools to maintain multiple simultaneous “relationships” across language barriers, producing fluent, emotionally resonant messages at scale. Researchers have observed Southeast Asian crime syndicates adopting AI-assisted face-swapping technology for video calls, further reinforcing false identities. The combination of emotional manipulation and increasingly convincing technology makes these scams extraordinarily difficult to detect, even for cautious individuals.

QR Code Phishing (“Quishing”) and the Mobile Security Gap

QR codes have become so embedded in everyday life — on restaurant menus, parking meters, transit stations, and business correspondence — that most people scan them without a second thought. Cybercriminals have systematically exploited this behavioral pattern in a growing category of fraud called quishing, short for QR code phishing. Because a QR code encodes its destination link as an image rather than visible text, it bypasses the URL-filtering tools built into most email security platforms. The attack shifts from a corporate security environment to the victim’s personal mobile device, where far fewer protections exist.

Keepnet Labs reported a fivefold increase in QR-based phishing attacks from 2024 to 2025. By 2025, 12 percent of all phishing attacks contained a QR code, and 68 percent of quishing attacks specifically targeted mobile users, according to Keepnet’s published statistics. The FBI’s Internet Crime Complaint Center issued its first formal public service announcement about QR code tampering as early as January 2022, warning that criminals were placing fraudulent QR code stickers over legitimate codes on payment systems and public signage. The tactic has since expanded considerably. Physical sticker attacks across 200 store locations in a documented 2025 incident caused $2.3 million in damage control costs alone, according to data cited by East Carolina Criminal University’s cybersecurity researchers.

In 2025 and into 2026, attackers began pairing quishing with AI-generated follow-up communications — after a victim scans a malicious code, they receive a convincing automated text or email nudging them to “complete verification” or “resolve an account issue,” increasing the likelihood of credential theft or financial loss. A NordVPN survey found that 73 percent of Americans scan QR codes without verifying where the code will take them first. Microsoft’s 2025 security report noted that QR-based delivery had become a significant vector in successful phishing campaigns that bypassed multi-factor authentication.

AI-Powered Phishing and Hyper-Personalized Social Engineering Attacks

Traditional phishing emails were often easy to identify: poor grammar, generic salutations, implausible sender addresses. Generative AI has largely eliminated these tells. AI-generated phishing emails now achieve click-through rates more than four times higher than their manually crafted counterparts, according to data cited by Vectra AI. Fraudsters compile personal data from previous breaches, public social media profiles, and commercially available data sets to construct messages that reference a victim’s employer, colleagues, recent purchases, or travel history — a technique known as spear-phishing. Guardio, a browser security firm, reported filtering one and a half times more scam and spam messages in November 2025 than in the same month in 2024, projecting that figure to nearly double by the end of 2026.

The World Economic Forum’s Global Cybersecurity Outlook 2026 found that 73 percent of organizations were directly affected by cyber-enabled fraud in 2025. Business email compromise (BEC) — in which attackers impersonate executives or trusted vendors to authorize fraudulent wire transfers — has been significantly amplified by AI. Fraudsters now deploy AI to clone writing styles from genuine corporate email threads, making impersonation nearly undetectable. Former Interpol Director of Cybercrime Craig Jones described the situation bluntly in public statements in early 2026, saying that AI has industrialized cybercrime, making attacks faster, more scalable, and harder to attribute.

A particularly dangerous evolution documented in 2026 involves self-modifying malware that uses large language models to mutate its own code in real-time as it spreads, rendering it nearly invisible to conventional antivirus tools. This category of AI-assisted malware represents what some security analysts have described as a fifth wave of cybercrime, operating at a speed and scale that outpaces the ability of traditional reactive defenses to respond effectively.

The Broader Technology Scam Landscape: Additional Threats to Watch

Beyond the headline-generating categories above, several additional scam types are expanding rapidly through digital channels in 2026. Sumsub’s 2025–2026 Identity Fraud Report documented a 180 percent increase in “sophisticated fraud” — a category defined by enhanced deception techniques, social engineering, and AI-generated synthetic identities — compared to 2024. Deepfakes now account for 11 percent of global fraudulent activity, according to Sumsub’s research. The fraud ecosystem has diversified significantly, with criminals targeting mobile payment systems, fake employment opportunities, and government benefit impersonation schemes.

Kaspersky warned in late 2025 that telecommunications-related scams are set to rise further in 2026 as the rollout of newer network technologies introduces fresh operational risks. The FTC’s Consumer Sentinel Network classified imposter scams as its top fraud category in 2025, with reported cases approaching one million annually and losses exceeding $3.5 billion in that category. Impersonation fraud cases rose approximately 19 percent year-over-year, driven primarily by AI-assisted personalization that makes fraudulent contacts significantly harder to dismiss.

What Dangerous Scams Spreading Through Technology Mean for Protection Strategies

The legislative response to AI-enabled fraud is accelerating. As of April 2026, the U.S. Senate Commerce Committee and the House Energy and Commerce Committee announced hearings focused specifically on AI voice fraud. Two significant legislative efforts are in progress: the Voice Cloning Protection Act, which would require explicit consent before a person’s voice can be used to train AI models or generate synthetic speech; and the DEEPFAKES Accountability Act, which would require digital watermarking of AI-generated audio and video to make synthetic media identifiable to detection tools. The FTC has also proposed expanded authority to regulate deceptive AI-generated communications under existing consumer protection mandates.

On a practical level, security experts recommend several layers of protection. Establishing a personal “safe word” with family members — a code that can be requested during any unexpected call claiming an emergency — is one of the simplest defenses against AI voice cloning impersonation. For financial transactions, verifying any unexpected wire transfer request through a separate, pre-established communication channel is essential, even if the requesting party’s voice or video appears genuine. The FBI explicitly advises that organizations can no longer rely on caller ID verification, as attackers routinely spoof numbers matching known contacts. Reporting suspected fraud to the FTC at ReportFraud.ftc.gov and the FBI’s IC3 at ic3.gov remains the recommended first step, and prompt reporting to financial institutions can, in some cases, enable account freezes before funds are fully transferred.

Frequently Asked Questions About Technology Scams

Staying One Step Ahead of the Fraud Wave

The dangerous scams spreading through technology in 2026 are not simply more frequent versions of the fraud that existed a decade ago — they represent a qualitative shift in what criminals can achieve with accessible, low-cost AI tools. A voice can be cloned from a birthday video. A cryptocurrency trading platform can display convincing returns until the moment it vanishes. A QR code printed over a legitimate one at a parking meter can silently redirect a payment. Awareness remains the most durable layer of defense, but it must be paired with verification habits that account for the fact that digital content — voices, video, and text — can no longer be trusted at face value. As regulatory frameworks catch up and detection technology matures, the responsibility for protection in the near term rests substantially with informed individuals who understand what these scams look like, how they operate, and where to report them when they appear.