Russian AI Call-Blocking Platform Now Handles 200 Million Calls a Day

Powered by artificial intelligence, the system identifies and blocks suspicious calls in half a minute — 30 percent faster than last year.

A Russian cybersecurity platform is processing an astonishing 200 million phone calls every day, filtering out spam and fraud in real time. Built by the telecom giant MTS, the service — called Defender — uses artificial intelligence to spot and shut down suspicious calls within 30 seconds. Engineers recently retrained the system’s model, making it 30 percent faster at recognizing and blocking unwanted calls, especially robocalls and telemarketing attempts.

The AI doesn’t just look at caller IDs. Each phone number is analyzed across more than 200 parameters: the volume of calls, the ratio of incoming to outgoing connections, and how often calls are successfully completed. During every live conversation, three neural models run in parallel, crunching more than a thousand parameters per second. According to MTS, the system now achieves a 95.17 percent accuracy rate in flagging fraudulent activity.

Between January and June 2025 alone, Defender blocked 1.5 billion calls from scammers and spammers — a 6 percent increase compared to the same period last year. The company claims the service has safeguarded both personal data and money for millions of subscribers.

Meanwhile, another Russian mobile operator has rolled out a tool that alerts users when their personal data leaks online, giving them the option to remove it from open sources before it spreads further.