MUMBAI, India, Oct. 5 -- Intellectual Property India has published a patent application (202621099238 A) filed by Dr. Anita Mahajan on August 17, 2026, for Ai-Driven Detection Of Injection Attacks In Api’s Using Bidirectional Recurrent Neural Networks.
Inventors include Dr. Ajay Varma; Dr. Shiv Shankar Rajput; Dr. Bharti Bhattad; Prof. Ritika Bhatt; Dr. Leeladhar Chourasiya; Dr. Nitin Kulkarni; and Prof. Krupi Saraf.
The application for the patent was published on October 02, 2026, under issue no. 40/2026.
Abstract: Injection attacks pose a significant risk to system security and user privacy due to the increasing reliance on application programming interfaces (APIs) for data transmission and reception. Approximately 25% of all vulnerabilities in online applications are attributed to injection attacks, according to recent statistics. The significance of employing trustworthy detection methods becomes evident in this context. The majority of the conventional attack detection methods rely on SQL injection attacks. Because of this, malicious actors are free to utilize injections in forms other than XML and JSON without fear of detection. The authors propose Bidirectional Recurrent Neural Networks (RNNs) as a novel approach to improve injection type discrimination. The issues that were mentioned will be addressed by this approach. Findings from this study suggest that by simultaneously processing forward and backward data sequences, bidirectional recurrent neural networks (RNNs) could maximize feature extraction. This makes injection assaults much easier to detect. Accuracy, precision, and memory are the three areas where the proposed system excels above the status quo. In addition to being user-friendly, it allows you to take notes in real-time. By utilizing this comprehensive monitoring solution, businesses may significantly lessen their vulnerability to injection defects. Apps and user data are safeguarded in this way. Keywords: leaving other forms, such as XML and JSON injections, unmonitored and possibly exploitable by hostile actors
Disclaimer: Curated by HT Syndication.
