Author(s): Kale Guymon
Mentor(s): Sayeed Sajal
Institution UVU
This project will evaluate the threat posed by SQL injection attacks through insufficient public app security, which can compromise sensitive database information. Through this research, we will analyze how SQL injections function, how data is stored and retrieved in databases, and how AI/ML can render these attacks. We will examine how insecure web apps enable SQL injection attacks and propose best practices for creating AI/ML algorithms that prevent unauthorized access. Many applications fail to implement adequate security throughout, leaving databases vulnerable to breaches. As organizations increasingly host sensitive client data in cloud databases, securing user information has become critical, especially since these databases are accessed through public networks and are not restricted by geographic boundaries. Attackers often exploit improperly secured login mechanisms to inject malicious SQL code, bypass authentication, and gain unauthorized access to sensitive data. By understanding the structure and flow of data, we aim to develop strategies to protect information at its source. The methodology will include reviewing previous case studies on SQL injection attacks and analyzing real-world examples to pinpoint the root of the problem. A focus will be placed on how this information is stored, where vulnerabilities often occur, and how backend code through ML algorithms like KNN and Ransom Forest can be secured to prevent potential breaches. Additionally, we will explore the role of encryption in modern data storage as a potential solution if a database is breached. Through this research, we look to better protect against the threat posed by SQL injections in modern web apps and provide developers with actionable AI detection model solutions to keep sensitive information secure.