AI Scammers Now Trick Users Into Authorizing Fraudulent Crypto Transfers

Unlike traditional hacks that require overcoming blockchain security, these scams exploit human psychology

Reported losses from deepfake scams in 2026 have already surpassed last year’s total by 263%, according to TRM Labs, signaling a sharp rise in AI-driven fraud targeting cryptocurrency users. The blockchain intelligence firm notes that attackers are shifting tactics, focusing less on hacking wallets or exploiting code and more on manipulating individuals into voluntarily authorizing transactions. This trend reflects a broader evolution in crypto crime where social engineering, powered by generative AI, is becoming the primary tool for theft. The AI-in-Crime Adoption Index, introduced by TRM Labs, categorizes these schemes as a growing threat class where deception replaces technical breaches. Scammers use realistic audio and video deepfakes to impersonate trusted contacts, such as family members or financial advisors, convincing victims to send funds under false pretenses.

Unlike traditional hacks that require overcoming blockchain security, these scams exploit human psychology, making them harder to detect and prevent through technical means alone. How Deepfakes Are Bypassing Wallet Security These frauds often begin with a seemingly legitimate call or message featuring a cloned voice or likeness of someone the victim knows. The scammer creates urgency—claiming a medical emergency, legal issue, or investment opportunity—to pressure the target into acting quickly. Once the user approves a transaction through their wallet interface, the funds are transferred irreversibly, even though no private keys were stolen or networks compromised. TRM Labs emphasizes that the blockchain itself remains secure; the vulnerability lies in the user’s decision-making process. Can Users Protect Themselves Against AI-Powered Deception? Experts recommend adopting verification habits, such as confirming requests through independent channels before approving any transaction. Multi-signature wallets and transaction limits can also reduce risk by requiring additional approvals.

TRM Labs warns that as AI tools become more accessible and realistic, the success rate of these scams is likely to increase unless users and platforms adopt stronger safeguards. Education and real-time anomaly detection in wallet applications may play a key role in mitigation. Frequently Asked Questions What makes deepfake scams different from traditional crypto hacks? Unlike hacks that target software vulnerabilities or private keys, deepfake scams rely on impersonation and psychological manipulation to trick users into authorizing transfers themselves. Why are reported losses rising so rapidly in 2026? The surge is linked to the growing availability and realism of AI-generated audio and video, which make fraudulent identities more convincing and harder to distinguish from legitimate communications. Are blockchain networks still secure against these attacks? Yes, the underlying blockchain technology remains uncompromised; the breach occurs at the human layer, where users are deceived into approving fraudulent transactions.