Vulnerability Assessment and Remediation Using Advanced AI Analysis

Vulnerability Assessment and Remediation Using Advanced AI Analysis

연구 분야: Infrastructure

학회: 2024 12th International Japan-Africa Conference on Electronics, Communications, and Computations (JAC-ECC)

In the ever-evolving landscape of cybersecurity threats, traditional vulnerability scanning tools often operate in isolation, leading to inconsistent accuracy and incomplete assessments. This paper proposes a system that leverage multiple tools of vulnerability scanners and advanced AI analysis so we can generate a more precise vulnerability reports. Integrating the output reports of different scanners and utilizing Large Language Models (LLMs) that previously trained on standard cybersecurity frameworks, our system provides actionable recommendations that enhances the proactive threat detection through system logs analysis. The system has a user-friendly dashboard that allows secure access and interaction with the assessment results. System evaluations demonstrated the effectiveness of the proposed system, showing an improvement of 50% in vulnerability detection rates, a reduction in time spent at vulnerability resolution of 60%, and a 50% satisfaction for users compared to previous manual methods. Testing on five different vulnerable machines, our system has significantly outperformed traditional approaches. These results highlight the potential of integrating advanced AI analysis into vulnerability assessment and remediation processes.