Scholarlink

<p>The cyber threat landscape is evolving with increasing sophistication, making robust risk management strategies essential. Cyber insurance has emerged as a key mechanism for organizations to transfer risk. While not yet mandatory, recent regulatory trends are encouraging the adoption of multiple technologies, potentially paving the way for broader implementation. However, traditional cyber insurance underwriting still relies heavily on questionnaire-based risk assessments, lacking a truly data-driven approach. This gap has received limited attention in existing research. In this survey we examine data-driven methodologies for cyber insurance as a critical component of modern risk management. We systematically review the literature on data sources, data collection practices, risk calculation methods, and pricing strategies. Traditional actuarial approaches are analyzed and contrasted with emerging AI-enabled methods, including supervised learning models, predictive analytics, and simulation-based techniques that enhance the estimation of incident likelihood and severity. We identify key challenges, such as data quality, lack of standardization, constrained data sharing, and model interpretability, and outline opportunities for future work aimed at integrating empirical, automated analyses into underwriting and pricing workflows. Overall, this survey offers a structured overview of existing approaches and highlights directions for developing more data-driven and AI-supported approaches to cyber insurance risk management.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Risk management for cyber insurance: a survey for data-driven approaches with the use of AI

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月20日

查看全文
  • Antonios Paragioudakis,
  • Nikos Komninos,
  • Michail Smyrlis,
  • Georgios Spanoudakis,
  • Christos Kloukinas

摘要

The cyber threat landscape is evolving with increasing sophistication, making robust risk management strategies essential. Cyber insurance has emerged as a key mechanism for organizations to transfer risk. While not yet mandatory, recent regulatory trends are encouraging the adoption of multiple technologies, potentially paving the way for broader implementation. However, traditional cyber insurance underwriting still relies heavily on questionnaire-based risk assessments, lacking a truly data-driven approach. This gap has received limited attention in existing research. In this survey we examine data-driven methodologies for cyber insurance as a critical component of modern risk management. We systematically review the literature on data sources, data collection practices, risk calculation methods, and pricing strategies. Traditional actuarial approaches are analyzed and contrasted with emerging AI-enabled methods, including supervised learning models, predictive analytics, and simulation-based techniques that enhance the estimation of incident likelihood and severity. We identify key challenges, such as data quality, lack of standardization, constrained data sharing, and model interpretability, and outline opportunities for future work aimed at integrating empirical, automated analyses into underwriting and pricing workflows. Overall, this survey offers a structured overview of existing approaches and highlights directions for developing more data-driven and AI-supported approaches to cyber insurance risk management.