AI-driven brand imitation scams surge in Serbia
Cybercriminals are using AI to imitate brand identities and conduct fraud at scale in Serbia, where 33.2 percent of fraud is now related to brand imitation. Research from Kaspersky shows that many victims are defrauded multiple times, causing financial harm and widespread loss of consumer trust in digital communications.
What happened
According to Kaspersky research, cybercriminals using AI are damaging brand reputation and user trust by imitating brand identities at scale. In Serbia, 33.2 percent of fraud is related to brand imitation using AI, and brand imitation is now among the three most common types of fraud via communication platforms worldwide.
Cybercriminals have become skilled at replicating brand identities and using AI to conduct fraud at scale, making it difficult for users to distinguish legitimate from fake brand communications. Many victims are defrauded multiple times, with some defrauded up to five times, and nearly one-sixth of affected consumers report being defrauded three or more times.
Reported impact
Beyond financial and emotional consequences for individuals, these scams cause long-term damage to relationships between users and brands and their interaction with brands. An overwhelming majority (99 percent) of consumers report losing trust in all incoming notifications via communication channels after being victimized, indicating near-total loss of confidence in the digital communication channels brands legitimately use to contact users.
The frequency of victimization maintains a vicious cycle in which distrust continues, as victims remain hypervigilant and skeptical. Users receive enormous volumes of messages, many designed to appear credible and create a sense of urgency, creating inevitable security gaps that allow fake messages to pass unnoticed and successfully persuade people to surrender personal data and money.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.