MUMBAI, India, Sept. 28 -- Intellectual Property India has published a patent application (202641112856 A) filed by Dr. Ramesh Shahabadkar on September 21, 2026, for Ai-Based Synthetic Cyberattack Generation And Machine Learning System For Automated Vulnerability Assessment.
Inventors include Dr. Ramesh Shahabadkar; Dr. Nandeeshwar S B; Dr. Krutika Ramesh Shahabadkar; Pallavi K V; Dr. Niranjan Kumar; and Dayanand Sagar Academy Of Technology And Management, Bengaluru.
The application for the patent was published on September 25, 2026, under issue no. 39/2026.
Abstract: The present invention relates to an artificial intelligence-based cybersecurity system configured to generate controlled synthetic cyberattack scenarios and automatically assess vulnerabilities of a target computing environment without requiring deployment of uncontrolled real-world attacks. The system integrates an artificial intelligence engine, synthetic attack generation module, machine-learning-based vulnerability assessment module, cyber environment modeling module, behavioral analysis module, risk scoring module, and adaptive learning module. The system receives authorized security parameters, network configuration characteristics, asset information, system behavior indicators, security policies, and historical security observations associated with a protected computing environment. Based on the received information, the synthetic attack generation module generates controlled and parameterized representations of cyberattack scenarios corresponding to different threat behaviors. The generated scenarios are executed or simulated within an isolated testing environment, thereby enabling security assessment while reducing the possibility of disruption to operational systems. Observed system responses, event patterns, security alerts, and vulnerability indicators are processed by the machine- learning module to identify abnormal behavior, estimate vulnerability conditions, classify security weaknesses, and determine corresponding risk levels. A vulnerability assessment engine generates an automated vulnerability profile comprising vulnerability categories, affected assets, severity scores, confidence values, and recommended defensive priorities. An adaptive learning mechanism uses assessment outcomes and subsequent security observations to update the synthetic scenario generation parameters and machine-learning models. The system thereby establishes a closed-loop cybersecurity assessment architecture in which synthetic threat generation, controlled testing, machine-learning analysis, vulnerability identification, risk prioritization, and adaptive model updating are integrated into a continuous process. The invention is applicable to enterprise networks, cloud environments, Internet-of-Things systems, industrial information systems, educational networks, data centers, and other authorized computing environments requiring automated and repeatable cybersecurity assessment.
Disclaimer: Curated by HT Syndication.
