SecurityWeek reports that Amazon Web Services has been averting 182,000 new malicious domains daily on average with its Mithra neural network graph model.
With 3.5 billion nodes and 48 billion edges, Mithra not only enables scoring of domain names but also advanced prediction of malicious domains up to months before their appearance in third-party threat intelligence feeds, according to AWS Chief Information Security Officer CJ Moses. "By assigning a reputation score that ranks every domain name queried within AWS on a daily basis, Mithra’s algorithms help AWS rely less on third parties for detecting emerging threats, and instead generate better knowledge, produced more quickly than would be possible if we used a third party," Moses added. Along with Mithra, AWS also touted its MadPot internal threat intelligence decoy system with advanced sensors and automated response features that have enabled successful blocking of threats from Sandworm, Volt Typhoon, and other state-sponsored advanced persistent threat operations.
