The telecom operator Beeline has launched a second AI model to combat phone fraud, this time aimed at protecting teenagers. As reported to TASS by the company's press service, "Cybergrandson" is deployed on the operator's own facilities and operates on the same principle as "Cybergrandmother," launched in 2025. The AI analyzes the caller's speech in real time, generates a response in the context of the conversation, and converts it into spoken language.
Statistics confirm the effectiveness of the approach. Since the beginning of 2026, "Cybergrandmother" has conversed with fraudsters for about 70 hours, with the maximum duration of a single call reaching 30 minutes. Approximately 500 fraudulent calls are redirected to the AI monthly. "Cybergrandson" extends this toolkit to a new audience – teenagers, who, after the elderly, are in a high-risk group.
Beeline announced the expansion of the cyberfamily so that in any call, the attacker would encounter an interlocutor matching the gender and age of the real number owner. Thus, the operator is effectively building a conveyor belt for utilizing fraudulent traffic: while the fraudster spends resources talking to a bot, real subscribers remain safe.
Комментарии