Symmetry Systems Powers Ahead with New AI-Powered Classification Features and Open Source Taxonomy

Symmetry unveils comprehensive AI-powered features and taxonomy integrating sensitive data identifiers, data types, and privacy frameworks; plans open source release to drive vendor consistency and industry collaboration.

SAN MATEO, Calif., Jan. 30, 2026 /PRNewswire/ -- Symmetry Systems, the Data+AI security company, today unveiled its AI-Powered Classification taxonomy that fundamentally transforms how organizations classify and secure sensitive data, as well as new capabilities for large enterprises to "Bring Your Own AI".

A Unified Taxonomy for Modern Data+AI Security 

The comprehensive taxonomy and Symmetry DataGuard itself supports classification and categorization of data with over 400+ sensitive data identifiers (PII, PHI, PCI, financial data, IP, credentials, etc), 500+ semantic data types (categories including contracts, board documents, financial records, healthcare documents, legal filings), regulatory classifications (GDPR, CCPA, HIPAA, PCI DSS, SOC 2, emerging AI governance, etc), and privacy data elements integrating into a unified model as the backbone for modern Data+AI Security programs. The company also announced plans to open source the taxonomy to drive industry-wide standardization and evaluations.

"The industry desperately needs open standards for data classification and categorization," said Sameer Sait, Sr Director, Information Security at Stanley 1913. "Vendor-specific taxonomies force organizations to maintain multiple overlapping frameworks and create unnecessary friction in security programs. Symmetry's vision for an open, standards-based taxonomy and supporting test data addresses a fundamental problem the entire industry faces. This is exactly the kind of leadership and collaboration we need."

In a significant step toward industry collaboration, Symmetry has also integrated the Fides privacy-by-code taxonomy into it's broader taxonomy and plans to release the combined taxonomy and supporting corpus of data, as an open source project. This addresses a critical gap in the data security industry: the lack of consistent classification standards across vendors, which forces organizations to maintain multiple mapping frameworks and slows adoption of comprehensive data protection programs.

The new taxonomy is available now to Symmetry Systems customers.

The Vision: Creating Open Source Industry Standards

"The data security industry has a taxonomy problem," said Dr. Mohit Tiwari, CEO of Symmetry Systems. "Every vendor has their own classification scheme, sensitive data identifiers, regulatory mappings, and policy language. Organizations waste resources translating between incompatible approaches instead of securing data."

"Think of this as the PyTorch moment for data security," added Dr. Tiwari. "PyTorch succeeded because it gave AI/ML domain experts a compact specification layer to define models, while abstracting away implementation complexity – the same model specification can target GPUs, TPUs, or CPUs without rewriting. Similarly, an open taxonomy and policy language lets privacy and data security experts express compact policies once, then automatically generate implementations targeting Unity Catalog in Databricks, IAM policies in AWS, OPA rules in Kubernetes, or access controls in Snowflake. Write the policy once in simple understandable terms, deploy it everywhere as technical enforcement."

The vision extends beyond classification to bridging business policy and technical enforcement. When a CPO says "vendors cannot access customer data," that should automatically generate Databricks Unity Catalog permissions, Open Policy Agent (OPA) policies, DLP policies, network segmentation rules, and audit configurations. Today, organizations manually translate requirements across disconnected systems. An open standard taxonomy provides the common language and technical mapping layer that makes this policy-as-code translation work at scale. By open sourcing the taxonomy and leveraging the well established privacy engineering taxonomy by Fides, Symmetry aims to enable vendor interoperability, accelerate compliance with automated policy generation, foster community innovation through benchmark datasets, and drive AI governance with common classification language – creating reproducible performance testing just as the AI community benchmarks models.

"Pytorch enabled rigorous performance testing – benchmarking which models excel at specific tasks and driving the entire field forward through reproducible results. Data security needs the same approach," Dr. Tiwari continued. "We're creating a comprehensive, open taxonomy integrating frameworks, with evaluation datasets so the community can test, benchmark, and continuously improve classification and categorization accuracy. This isn't about competitive advantage – it's about enabling the rigorous evaluation and continuous improvement that drove AI forward."

The company will announce timing, governance structure, and initial benchmark datasets in the coming weeks.

"Bring Your Own AI" Infrastructure Flexibility 

Complementing the new taxonomy, Symmetry announced expanded "Bring Your Own AI" (BYOAI) infrastructure support, allowing customers to power their classification using their own enterprise AI subscriptions (Azure OpenAI, AWS Bedrock, Google Vertex AI), or their own private GPU infrastructure for data sovereignty. This eliminates vendor lock-in and allows organizations to apply the industry-standard taxonomy using their own AI infrastructure that meets their specific security and compliance requirements.

These capabilities reinforce Symmetry's unique position as the only Data+AI security platform delivering comprehensive classification without requiring data to leave their custody and control.

Unlike cloud-dependent Data Security Posture Management solutions that require data to flow through vendor infrastructure, Symmetry's architecture can bring the taxonomy and AI-powered classification to wherever data lives – delivering modern AI intelligence without the security and compliance compromises of cloud-only approaches.

The platform enables security and data teams to apply consistent classification using a unified taxonomy across all data systems, deploy in airgapped environments with full AI-powered functionality, leverage existing AI investments, maintain complete data sovereignty, accelerate compliance reporting with built-in regulatory mappings, enable responsible AI training with consistent taxonomy, and prepare for open standards designed to incorporate community contributions.

The BYOAI infrastructure support is available now to Symmetry Systems customers. Organizations can learn more and request a demo at https://www.symmetry-systems.com/get-started.

Press Contact

Claude Mandy
Symmetry Systems
[email protected]

About Symmetry Systems 

Symmetry Systems is the Data+AI security company, providing organizations with the industry's only comprehensive Data + AI Security Platform that discovers, classifies, protects, and monitors sensitive data across. Born from award-winning DARPA-funded research at UT Austin, our AI-powered platform delivers comprehensive Data+Ai security across all major cloud environments, SaaS applications, on-premise data stores, legacy systems, and airgapped environments. Our "get everywhere" philosophy continuously expands connector coverage to secure data wherever it lives—in all major cloud environments, SaaS applications, and on-premise data stores-including mainframes, legacy systems and airgapped environments

By uniquely merging both identity and data context, Symmetry provides what other DSPM vendors cannot: complete visibility where data exposure meets agentic identities. Organizations use our platform to eliminate unnecessary data, remove excessive permissions, accelerate compliance and cloud migration, and reduce attack surfaces – while safely enabling agentic AI systems with the identity-aware data context they require.

Innovate with confidence with Symmetry Systems. 

SOURCE Symmetry Systems, Inc