MUMBAI, India, Aug. 12 -- Intellectual Property India has published a patent application (202641091996 A) filed by Ifet College Of Engineering on July 29, 2026, for Standalone Prompt Manipulation Detection System For Identifying Coordinated Bias Injection.
Inventors include Dr. Rosario Gilmary; Dr. R. Thendral; Mr. M. Arunkumar; and Mrs. N. Manvizhi.
The application for the patent was published on August 07, 2026, under issue no. 32/2026.
Abstract: The present invention relates to a standalone prompt manipulation detection system (100) for identifying coordinated prompt manipulation associated with bias injection influencing generative artificial intelligence environments. The system (100) comprises a standalone prompt monitoring controller (102) operatively coupled to a prompt acquisition interface (104), a prompt sequence encoding module (106), a prompt pattern analysis module (108), a coordinated bias detection module (110), an alert generation module (112), a communication interface (114), and a power supply module (116). The prompt acquisition interface (104) receives prompt-related digital communication inputs from one or more communication sources, and the prompt sequence encoding module (106) generates structured prompt sequences for evaluation. The prompt pattern analysis module (108) evaluates the structured prompt sequences to identify coordinated prompt patterns, while the coordinated bias detection module (110) identifies prompt manipulation associated with bias injection and generates detection outputs through the alert generation module (112).
Disclaimer: Curated by HT Syndication.
