MUMBAI, India, Oct. 12 -- Intellectual Property India has published a patent application (202621088303 A) filed by Persistent Systems Limited on July 20, 2026, for System And Method For Explainable Decision Intelligence For Ai Workflow Execution.
Inventors include Mr. Nitish Shrivastava; and Mr. Pradeep Kumar Sharma.
The application for the patent was published on October 09, 2026, under issue no. 41/2026.
Abstract: A system (100) captures machine-readable provenance of decisions made during execution of a multi-model artificial-intelligence workflow and generates faithful, role- appropriate explanations. A decision-intelligence recorder (110) creates a decision intelligence record (200), capturing candidate actions, eligibility and rejection reasons, predicted outcomes, constraints, policies, selected actions, uncertainty, resource impact, evidence references, and version identifiers, which are stored in a DIR store (120). A provenance graph builder (130) generates a provenance graph (250) linking evidence, model invocations, verification, policy evaluations, and outcomes. An attribution engine (140), counterfactual engine (150), and reason-code generator (160) produce reason codes, metric attributions, and feasible counterfactuals. A faithfulness verifier (170) validates explanations, while a role-based redaction/render module (180) controls disclosure. A tamper-evident log (195) and audit/replay module (190) enable reproducible auditing independent of the optimizer.
Disclaimer: Curated by HT Syndication.
