Cloud security and compliance provider for infrastructure and applications, Threat Stack announced that it combines telemetry collection, security rules, human expertise, and machine learning to allow customers to detect known and unknown risk with context. Its new machine learning ThreatML enhances security observability for the Threat Stack Cloud Security Platform, Threat Stack Oversight, and Threat Stack Insight with anomaly detection.
According to the announcement, ThreatML leverages Threat Stack Cloud Security Platform’s over 60 billion events per day, which is collected, normalized, and analyzed from customer cloud infrastructure and applications to train its machine learning models and detect anomalous behavior. Threat Stack’s rules engine and advanced machine learning capabilities allow detecting, prioritizing, and responding to known and unknown threats quickly. Combining full-stack telemetry, machine learning, rules, and human expertise accelerate mean-time-to-know, focus on high-severity threats, save time, and reduce cost. Brian Ahern, CEO, Threat Stack, said,
“Machine learning is often promoted as a silver bullet solution to all problems. With the introduction of ThreatML we are combining the industry’s best security telemetry, rules engine, human expertise, and now machine learning to create a truly powerful cloud security solution capable of detecting known and unknown risks. This provides our customers with better security coverage, unparalleled contextual findings, and cost benefits by reducing mean time to know and respond to threats.”