Artificial intelligence can now clone a person's voice using just seconds of audio, a capability that scammers are exploiting to perpetrate impostor fraud, according to the University of Cincinnati's Kimberly Hyun. The technology allows fraudsters to mimic loved ones or authority figures convincingly.
This development comes as impostor scams remain one of the most common forms of fraud, according to the Federal Trade Commission. The ease of voice duplication lowers the barrier for criminals, who previously needed more time or access to pull off such schemes.
Hyun's research highlights that the AI models require minimal data — a short clip from a social media video or a phone call is sufficient to generate a convincing replica. No specialized equipment or technical expertise is needed to deploy the tools.
The implications are significant for both consumers and businesses. Individuals may receive calls that sound exactly like a family member asking for money, while companies could face deepfake-based social engineering attacks targeting executives.
Experts recommend verifying unexpected requests through a second channel, such as a text message or in-person check, before taking action.