MUMBAI, India, Nov. 2 -- Intellectual Property India has published a patent application (202621105673 A) filed by Nilesh Shrikrishan Aswar; Dr. Ganapati Appasaheb Patil; and Dr. Raman Bane on September 02, 2026, for System For Detecting, Classifying, And Iteratively Mitigating Security Vulnerabilities In Artificial Intelligence Generated Source Code.
Inventors include Nilesh Shrikrishan Aswar; Dr. Ganapati Appasaheb Patil; and Dr. Raman Bane.
The application for the patent was published on October 30, 2026, under issue no. 44/2026.
Abstract: ABSTRACT A computer implemented system and method detect, classify, analyse, and mitigate vulnerabilities within code generated by artificial intelligence systems. A data collection module obtains generated code and metadata, while a sampling module selects a statistically representative sample according to population size, confidence level, and margin of error. A vulnerability detection module scans selected code using static application security testing, dynamic application security testing, and software composition analysis engines. Detected findings are normalized and classified according to security taxonomies, Common Weakness Enumeration identifiers, severity scores, and compliance frameworks. Statistical analysis identifies recurring weaknesses, high risk programming languages, model related patterns, and relationships between code characteristics and vulnerability severity. A mitigation module converts findings into development guidance, testing configurations, and augmented prompts containing vulnerability specific remediation instructions. An artificial intelligence code generator uses each augmented prompt to regenerate hardened code. The regenerated code is rescanned, compared with the original findings, and repeatedly corrected when acceptance criteria remain unmet. An integrated development environment plugin provides real time inline warnings and automated remediation controls. Lifecycle integration applies safeguards during requirement analysis, design, building, testing, deployment, and maintenance. The invention thereby delivers continuous, measurable, compliance aware DevSecOps protection for artificial intelligence assisted software development environments.
Disclaimer: Curated by HT Syndication.
