Russia is investing 2.27 billion rubles to deploy machine-learning filters within the national network filtering system to identify and block VPN connections. The predictive model analyzes packet sizes, transmission intervals, session lengths, and server patterns without attempting to decrypt user content. Telecom sources say the traffic-analysis algorithm misclassifies traffic in 5% to 6% of cases, while corporate VPNs and cloud applications may also be disrupted because hosted subnets can be blocked.
Source:TechRadar · Read Original
Leave a Reply