At the core of empow’s technology is an AI based correlation engine that leverages Belief and Bayesian models to automatically correlate logs, including through cause-and-effect relationships. This automates the process of root-cause and triage analysis as well as predicting the attacker’s next steps, and clustering the logs into only a few real attack stories.  The result is a graphical time-based representation of real attacks and prioritized entities that are at top risk.

empow’s NLP-based threat classification center continuously collects threat intelligence from multiple commercial and open threat centers and enriches the logs with it.  It then uses the power of NLP to automatically and continuously classify the logs into potential attack behaviors represented by attack techniques, tactics, malicious software and other parameters, representing the logs one language of ATT&CK.
Read more in our AI White Paper.

User activity logs are processed by empow’s UEBA, which establishes normal behavioral profile patters and flags deviations which reflect potential attack behaviors.
Read more in our UEBA White Paper.

empow’s DPI network sensors feed the network traffic analytics module, which learns the normal traffic patterns between hosts within the organization and flags anomalies that reflect both known and unknown potential attack behaviors.
Read more in our NTA White Paper.

empow’s EDR agent is based on the Elastic agent which supports malware prevention, exploit detection, unusual processes, user and admin activities, operational disruption, misuse of network protocols and reach host telemetry collection including: DLL and driver load, process status change, registry, files, DNS, network and security events. Detections are based on machine learning and rules which are customizable.

SOAR – or predictive response – capabilities are enabled by empow’s unique automated technologies which automate classification and prioritization of threats and remediation processes. 16 patents form the basis for these capabilities, utilizing a number of technology approaches including: Artificial Intelligence (AI) Natural Language Processing (NLP) and Belief and Bayesian Networks (BBN) algorithms, and reinforced with User Entity Behavior (UEBA), Network Traffic Analysis (NTA), and Threat Intelligence (TI) engines. Together these algorithms and technologies enable i-XDR to find attacker “intent” before the full attack is carried out and execute predictive response to prevent it.