MUMBAI, India, Oct. 26 -- Intellectual Property India has published a patent application (202621107585 A) filed by Dr. Afreen Banu; Mrs. Shivani Hemraj Patil; Ms. Prachi Wagde; Mrs. Vaishnavi Vinit Mestry; Mrs. Vaishali Bhanudas Kosamkar; and Mrs. Prithi Kanaga Paul on September 07, 2026, for An Ai-Based System And Method For Automated Privacy Risk Detection And Adaptive Data Protection.
Inventors include Dr. Afreen Banu; Mrs. Shivani Hemraj Patil; Ms. Prachi Wagde; Mrs. Vaishnavi Vinit Mestry; Mrs. Vaishali Bhanudas Kosamkar; and Mrs. Prithi Kanaga Paul.
The application for the patent was published on October 23, 2026, under issue no. 43/2026.
Abstract: The present invention provides an AI-based system and method for automated privacy risk detection and adaptive data protection in heterogeneous computing environments. The system semantically discovers privacy-relevant data, maps processing context, and maintains a privacy knowledge graph representing relationships among data elements, subjects or subject classes, purposes, consent or authorisation states, recipients, locations, retention states and protection actions. An artificial- intelligence privacy-risk inference engine processes a contextual feature vector including semantic and graph-derived features to determine an initial privacy-risk state and counterfactual residual-risk estimates. A constraint engine independently enforces mandatory purpose, consent, retention, recipient and transfer constraints. An adaptive protection orchestrator selects and executes a minimum sufficient protection action predicted to reduce risk to a target residual-risk threshold while preserving permitted processing utility. The protected data is re-evaluated to determine observed residual risk, and protection is automatically escalated when the threshold is not satisfied. Verified outcomes are recorded as tamper-evident evidence and may recalibrate learnable inference or action-selection parameters without modifying mandatory constraints. Accompanied Drawing [FIGS. 1-2].
Disclaimer: Curated by HT Syndication.
