AT&T Specialist from India Secures Two Key US Patents for Q&A Systems and Online Fraud Detection

Japa Sharath Reddy, an Indian specialist employed by the telecommunications giant AT&T, has been awarded two distinct patents by the United States Patent and Trademark Office (USPTO) spanning late 2024 and early 2025. These accolades recognize Reddy’s significant contributions across the domains of information retrieval and cybersecurity. AT&T, a company whose legacy of innovation stretches back to Alexander Graham Bell's invention of the telephone in 1876, is vigorously safeguarding its intellectual property. In 2024 alone, the company concentrated approximately 99% of its patent applications and secured 97% of its granted patents within the United States.

The first patent, officially granted on December 10, 2024, details the creation of a hybrid question-and-answer system that leverages a knowledge graph model for organizing complex information. Knowledge graphs are fundamental to many artificial intelligence applications, including advanced search and Q&A functionalities, enabling more effective extraction and processing of intricate data sets. Such graph-based structures, when built upon patent data, can aggregate facts to support logical inference and targeted searching in complex engineering scenarios.

The second patent, issued on March 4, 2025, focuses on an innovative methodology for identifying suspicious online activities by analyzing user behavioral patterns. This machine learning-driven approach allows for the detection of anomalies that might easily slip past traditional, rule-based analysis systems. In the realm of financial fraud detection, this pattern analysis—which considers transaction types and monetary thresholds—has proven highly effective, achieving accuracy rates exceeding 94% in certain established models.

Japa Sharath Reddy, who hails from Alugunur village within the Karimnagar Municipal Corporation, has demonstrated impressive inventive versatility by securing these two patents in such a short timeframe. His work at AT&T, where the firm utilizes over 600 machine learning and AI models, underscores the critical importance of intellectual property within the modern telecommunications sector. AT&T, which ranked sixth among US corporations for the volume of AI-related patents, is currently integrating generative AI to manage vast and complex data sets.

The innovation in fraud detection, specifically employing behavioral pattern analysis, aligns perfectly with broader industry shifts. Systems based on behavioral models can automatically identify emerging threats without constant manual rule updates. Earning two patents in such distinct yet interconnected fields—information structuring and behavioral security—is a testament to Reddy’s profound technical expertise. These achievements formally validate the worth of his developments, as the USPTO has acknowledged their novelty and utility in the ongoing evolution of data-driven and AI-powered services.