Enhancing Methods of Protection in the Amazon Web Services Cloud with Artificial Intelligence and Machine Learning

Abstract

Purpose: To investigate protection methods in the Amazon Web Services cloud environment using artificial intelligence and machine learning and to assess the effectiveness of AI-based security tools compared to traditional approaches.

Method: The study employs an experimental approach in the Amazon Web Services (AWS) cloud environment. Simulated cyberattacks, including unauthorized access, data exfiltration, web-based attacks, and privilege escalation, were performed to assess security effectiveness. A comparative analysis was conducted between traditional security mechanisms (CloudTrail, WAF) and AI-driven security tools (Amazon GuardDuty, Macie). The evaluation focused on detection accuracy, response time, and adaptability, reflecting the study’s findings how effectively each method detects and mitigates security threats in a cloud environment.

Findings: Security tools leveraging artificial intelligence, such as GuardDuty and Macie, provide more effective threat detection than traditional security methods. They demonstrate high accuracy, reduce false positives, and enable faster response to potential attacks.

Theoretical implications: The study deepens the understanding of AI's role in cloud security methods and highlights the need to integrate both traditional and automated security strategies.

Practical implications: The findings offer recommendations for implementing automated threat detection systems and improving security monitoring in the Amazon Web Services cloud environment.

Value: This research highlights the advantages of integrating artificial intelligence into cloud security and proposes practical solutions to enhance protection strategies.

Future research: Future studies may explore deep learning-based attack prediction, enhanced behavioral analytics, and the development of self-learning security systems.

Paper type: Conceptual research.