MUMBAI, India, July 24 -- Intellectual Property India has published a patent application (202631085265 A) filed by Dr. Anjan Saikia; Dr. Dipen Nath; Dr. Rupjyoti Dutta; Dr. Binod Chandra Borah; Sandeep Dwarkanath Pande; Dr. Sandip A. Kahate; Madhuri Navnath Gurav; Sajal Suhane; Dr. Devendra Kumar Singh; and Yogesh B. Pawar on July 12, 2026, for Artificial Intelligence Based System And Method For Providing Synchronized Identical Recommendation Information To A User Across Multiple Digital Interfaces.

Inventors include Dr. Anjan Saikia; Dr. Dipen Nath; Dr. Rupjyoti Dutta; Dr. Binod Chandra Borah; Sandeep Dwarkanath Pande; Dr. Sandip A. Kahate; Madhuri Navnath Gurav; Sajal Suhane; Dr. Devendra Kumar Singh; and Yogesh B. Pawar.

The application for the patent was published on July 17, 2026, under issue no. 29/2026.

Abstract: An artificial intelligence-based system and method for providing synchronized identical recommendation information across multiple digital interfaces are disclosed. The system identifies interfaces associated with a common user and generates a temporally fixed user- context snapshot. An artificial intelligence recommendation engine processes the snapshot once during a synchronized recommendation cycle. A resulting recommendation set is converted into a canonical recommendation object containing an immutable semantic payload, a recommendation version identifier and a content fingerprint. Interface-compatible representations are generated while preserving recommended entities, ranked order, action identifiers and validity conditions. A consistency validation engine verifies semantic correspondence between each representation and the canonical object. A synchronization controller coordinates release of validated representations according to interface readiness states. A delivery state ledger records presentation acknowledgements, while a cache state controller replaces outdated versions. Version-linked interaction events from different interfaces are reconciled into a common recommendation state, thereby preventing recommendation divergence and reducing repeated model execution. Accompanied Drawing [FIGS. 1-2]

Disclaimer: Curated by HT Syndication.