How It Works

Data Ingestion

• Collects vulnerability, asset, and exposure data from scanners (Qualys, Tenable, Rapid7, etc.).
• Ingests contextual metadata: network topology, exploit data, business criticality, and historical remediation patterns.

Feature Engineering

• Converts raw data into risk vectors, incorporating factors like:
• Asset value and sensitivity
• Exposure level (public, internal, isolated)
• CVSS base and temporal metrics
• Threat intelligence (CISA KEV, EPSS, MITRE ATT&CK)
• Remediation SLA performance and incident history

Machine Learning Model

• A hybrid model combines supervised learning (historical incident patterns) and unsupervised clustering (emerging risk correlations).
• Continuously retrains on live vulnerability outcomes improving prioritization accuracy over time.

Dynamic Risk Score Output

• Generates a normalized 0–500 score, mapped to Risk Grades (A–D).
• Scores update in real time when:
• Exploits become active in the wild
• Asset criticality changes
• A vulnerability is remediated or new dependencies are detected

Key Capabilities

Supported Tools:

• Predictive Prioritization: AI forecasts which vulnerabilities are most likely to be exploited in your environment.
• Contextual Weighting: Automatically assigns higher weight to vulnerabilities on high-value or externally exposed assets.
• Adaptive Learning: Improves accuracy over time using your organization’s historical risk and fix data.
• Exploit Intelligence Integration: Pulls data from global feeds like CISA KEV, NVD, and ExploitDB to adjust scores dynamically.
• Scenario Simulation: Allows “what-if” analysis e.g., what happens to organizational risk if a tier remains unpatched.

Benefits