Contemporary cyber threats are promoted by human actions as opposed to technology gaps. The conventional Security systems are not enough to combat the social engineering tactics and attack with user interaction. The proposed research is an AI-based Threat Detection and Response (TDR) system that is designed to identify malicious activity in an online transaction. The system Laura looks at behavioral surveillance indicators like the movements of the mouse, the dynamics of key strokes, rest intervals, form usage patterns as well as changes made when filling in the form. The processing of these Behavioral characteristics is done using the lightweight machine-learning algorithm to categorize the activities into normal and suspicious in real-time. The system has the capability of automatic responses to risk evaluation which can be warning to the user, block the form or can be a more intensive verification process. As proven by experimental findings, the AI-grounded TDR framework is more successful than the traditional rule-driven security systems because it foresights the threats, minimizes false positives, and offers mechanisms providing active defense. Behavioral Intelligence implementation coupled with real-time scalability is an efficient and practical way of dealing with the new cybersecurity threats.
R et al. (Thu,) studied this question.