[REVISION HISTORY]

Evolution of AI-driven deepfake scams and defensive tech

Updated 35 times since CLSTR started tracking revisions of this situation.

What changed

2026-10-04 06:43 UTC → 2026-10-08 13:15 UTC · added ~~removed~~

Cybersecurity authorities continue to warn of evolving deepfake scams using artificial intelligence to impersonate trusted individuals. In India, SEBI has warned of the ‘Boss Scam,’ where fraudsters use AI-generated video and voice cloning to impersonate executives. Similarly, the Bulgarian National Bank (BNB) has issued warnings regarding criminals cloning the voices and images of relatives or colleagues. In Europe, these crimes are estimated to cause losses of approximately 8,000 euros per minute. Criminal tactics are expanding to include ‘quishing’ (QR code phishing), where physical mail containing fake bank cards directs victims to fraudulent websites. Authorities in France and Portugal have issued alerts regarding these schemes. Furthermore, cybercriminals are using ‘silent calls’ to harvest data. Experts from Kaspersky note that these brief interactions allow attackers to confirm active numbers, identify when users are likely to answer, and collect audio samples to replicate voice biometrics for identity impersonation. In Latin America, data from the Who Calls app indicates that 88% of users have received unwanted calls, with approximately 11% linked to fraud attempts. AI-driven fraud is also increasingly targeting the real estate sector. Fraudsters utilize deepfake audio and AI-generated documents to impersonate property owners and professionals to intercept closing funds. According to the FBI’s 2025 Internet Crime Report, real estate fraud losses surpassed $275 million, up from $173 million in 2024. The agency also reported over 22,000 AI-related fraud complaints, with associated losses exceeding $893 million. Recent incidents highlight the scale of these threats. In Italy, a fraud targeting Fideuram bank used a fake WhatsApp message appearing to be from the CEO of Intesa Sanpaolo and an AI-generated voice mimicking a senior law firm partner, resulting in the transfer of approximately 95 million euros to overseas accounts. In Spain and Latin America, social engineering tactics are intensifying.

Versions

Only revisions since CLSTR began indexing content versions appear here. Select a version to see what changed compared to the one before it.