83 pc of AI voice scam victims in India suffer financial…

In January 2026, a school teacher named Smita, based in Indore, Madhya Pradesh, became a victim of a sophisticated AI-driven voice cloning scam. The incident highlights the growing threat of cyber-fraud in India, where attackers leverage artificial intelligence to mimic the voices of trusted family members and friends to facilitate financial theft.

What happened

Smita received a phone call from an individual whose voice was indistinguishable from that of her cousin. The caller claimed that a friend had suffered a heart attack and required immediate funds for medical treatment. Relying on the familiarity of the voice and the urgency of the situation, Smita transferred INR 97,500 across four separate transactions. The following morning, she discovered that her cousin had never placed the call. Police authorities investigating the matter suspect that fraudsters utilized AI-based voice-cloning technology to execute the deception. Experts suggest that such scams often rely on audio samples harvested from social media, YouTube videos, or podcasts to train AI models to replicate a specific person's pitch, accent, and cadence. By combining this technology with traditional social engineering tactics—specifically urgency, fear, and emotional manipulation—criminals are able to bypass standard security instincts. Bhavishya Pandit, a Senior Data Scientist, notes that voice is a foundational trust signal, making it particularly effective for exploitation. To mitigate these risks, experts recommend verifying urgent requests by disconnecting the call and reaching out to the individual through a known, independent channel or establishing a private family safe word.