Tuesday, July 1, 2025

Data Squared (Data²) Secures Patent for Hallucination-Resistant

Related stories

Brodeur Partners Unveils Relevance AI for Competitive Edge

New Suite of AI Services Combines Behavioral Science with...

Riskonnect Unveils AI Governance for Next-Gen Risk Management

New AI governance solution enables organizations to responsibly harness...

Danaher Names Martin Stumpe as Chief Technology & AI Officer

Danaher Corporation, a leading global life sciences and diagnostics...

databahn Unveils GenAI-Powered Sales Intelligence Tool

databahn announced the launch of its next-generation GenAI Sales...

VIDIZMO Named Microsoft Partner Across All Azure Areas

VIDIZMO secures Microsoft Solutions Partner status for Data &...
spot_imgspot_img

Data Squared (Data²), a Service-Disabled Veteran-Owned Small Business dedicated to helping public and private sector organizations unlock the full potential of their data, announced that the U.S. Patent and Trademark Office (USPTO) has issued a key patent directed to its groundbreaking Graph and Hybrid retrieval-augmented generation (RAG) based AI systems.

Data² is building one of the industry’s most advanced patent portfolios focused on hallucination-resistant, explainable AI (XAI). The company has established a new benchmark for trust, transparency, and performance in mission-critical environments.

“AI hallucinations and a lack of explainability have long been the Achilles’ heel of scaled AI adoption, especially in high-stakes environments,” said Jon Brewton, Founder & CEO of Data². “While traditional RAG systems retrieve and inject relevant external data into the generation process, our technologies go much further, making AI-generated answers fully transparent, verifiable, and grounded in traceable source data. We are committed to delivering solutions that empower decision makers to trust, trace, and verify every answer from any AI system.”

The Data² Technology

Unlike conventional RAG systems, Data²’s reView platform delivers dynamic, visual explainability, allowing users to trace every AI-generated answer back to its precise source data. The solution is LLM-agnostic, deployable in any environment – from cloud-native to fully air-gapped on-premises – and hardened with zero-trust security and enterprise grade encryption. This marks a step-change in AI capability and sets a new benchmark for trust and transparency in high-stakes industries.

Data²’s approach enables:

  • Hallucination Resistance: Every AI-generated answer is anchored in verifiable, traceable sources, defined through graph-connected evidence networks.
  • Full-Cycle Explainability: The system delivers detailed, visual traceability showing which data contributed to each response, complete with source attribution and data record-level citations.
  • Unified Data Integration: Structured, unstructured, and time series data are seamlessly integrated into harmonized data models.

Also Read: Datamaran Launches Core ESG & Reporting Intelligence Tool

Applications of Data²’s technology include:

  • Legacy Vendor Systems Integration
  • Offensive and Defensive Cyber
  • Unified All-Source Intelligence
  • Forensic Accounting
  • Investigation Case Management
  • Oil & Gas Asset Management
  • Secure Supply Chain Risk Management
  • Regulatory Analysis for Permitting
  • Contested Logistics Optimization
  • Legal Research and Case Analysis

Data² is actively licensing its systems to enterprise partners and embedding them into and along-side other existing commercial models and platforms. Purpose-built for environments where traceability and accountability are essential, these systems deliver regulatory-grade auditability, cross-modal data integration, and real-time explainability, while systematically mitigating hallucinated content in responses.

Source: Globenewswire

Subscribe

- Never miss a story with notifications


    Latest stories

    spot_img