Wiz Explains How Its AI Engine Finds and Contextualizes Exposed Sensitive Data
Summary
Wiz describes building an AI-powered data classification engine that analyzes files and their context to identify sensitive data. In a one-week test of public cloud storage, it reported more than 2,700 findings across hundreds of buckets.
Key points
- The system uses specialized AI agents and an iterative analysis pipeline to identify files and sensitive data that rule- and regex-based scanners may miss.
- It assesses file purpose, data categories, sensitivity, severity, and evidence, using context such as metadata, data volume, recency, and checks for synthetic or test data.
- Wiz says production optimizations include model selection by task, batching, budget controls, a sandbox, and grouping similar files to reduce redundant analysis.
- The system's findings feed into the Wiz Security Graph, alerts, and dashboards.
- Wiz says preview reviews against original files helped refine detection patterns, severity assessments, and confidence thresholds, bringing false positives near zero while maintaining high recall.
- In a one-week analysis of publicly accessible cloud storage, Wiz reported more than 2,700 sensitive-data findings across hundreds of buckets and nearly 2,000 files.
- Wiz says the engine now supports classification across cloud storage, cloud drives, endpoints, and attack simulation.
Article Details
- Topic
- AI-powered data classification and context engine for discovering and assessing sensitive data across cloud storage, SaaS, and endpoints.
Vendors
Products
Wiz DSPMDiscover how Wiz discovers, classifies, and contextualizes sensitive data risks across IaaS, SaaS, and virtual drives. Request a demo or explore Wiz DSPM.Wiz Security GraphSecurity teams get a complete picture of every finding. This output feeds directly into the Wiz Security Graph, triggers alerts, and populates dashboards.