AI-DRIVEN DETECTION OF INJECTION ATTACKS IN NEURAL API’S USING BIDIRECTIONAL RECURRENT NETWORKS

Abstract

The growing reliance on APIs for data transfer in online applications has made them a prime target for injection attacks, posing serious concerns to system integrity and user privacy. Recent figures show that injection attacks account for around 25% of all online application vulnerabilities, emphasizing the importance of robust detection techniques. Traditional detection systems generally focus on SQL injection attacks, leaving other forms, such as XML and JSON injections, unmonitored and possibly exploitable by hostile actors. To fill these shortcomings, this research suggests a unique methodology that uses Bidirectional Recurrent Neural Networks (RNNs), to improve the identification of different injection types. The work shows that bidirectional RNNs maximize feature extraction by examining data sequences in both forward and backward orientations, resulting in higher accuracy in identifying injection attacks. The suggested system beats existing methods in terms of accuracy, precision, and recall while also providing a user-friendly interface for real-time predictions. Implementing this comprehensive detection technology allows enterprises to substantially limit their susceptibility to injecting vulnerabilities, eventually protecting their apps and user data.