772 résultats
pour « Autre »
Cyber risk presents significant challenges to society, yet its statistical behavior remains insufficiently understood. This paper analyzes three databases to study cyber risk dynamics. It identifies increasing frequency and severity, particularly in malicious events since 2018. Persistent heavy-tailedness across risk categories implies lower insurance demand and potentially heightened risk levels for firms.
Effective transition to a sustainable economy requires robust emission reduction plans from companies, crucial for capital allocation and risk management. Transition disclosures serve as a market compass towards net-zero goals, mitigating financial risks. Despite various frameworks, inconsistencies persist, prompting a proposed set of 64 indicators for comprehensive assessment using natural language processing.
Major financial centers introduced varied climate disclosure rules, notably Scope 3 mandates. EU and California led with mandates, while the SEC proposed but later removed them in 2024. Challenges include accuracy, standardization, and compliance costs. EU provides institutional support, but the U.S. lacks it, raising reporting stakes.
“The financial impact of cybercrime paints a concerning picture. According to the FBI's Internet Crime Complaint Center (IC3), cybercrime complaints in 2023 reached record highs, with reported losses exceeding $10 billion (IC3, 2023). Furthermore, IBM's 2023 Cost of a Data Breach Report estimates the average global cost of a data breach to be a staggering $4.5 million (IBM, 2023). These statistics highlight the immense financial burden cybercrime places on individuals, organizations, and governments.”
The study presents an innovative approach for flood disaster risk assessment and ecological monitoring using Bayesian networks (BNs) and geographic information systems (GIS). By integrating diverse data sources, the BN-GIS model provides a holistic risk profile of flood-prone urban ecosystems in Yinchuan, China. This approach allows for mapping vulnerable hotspots, quantifying uncertainties, and informing sustainable urban planning and disaster resilience efforts.
The modern auditing process aligns with Bayesian methods, enabling auditors to use tools and techniques for targeted audit procedures and assurance modeling. This approach integrates sampling results and supports intuitive, easy-to-implement models for field use. The model generalizes the audit risk model, applying Bayesian techniques and the Stringer posterior.
“... we provide evidence that setting the risk-weighted ratio at 15\% and the simple leverage ratio at 10% would significantly decrease the probability of default without hampering banks' activities.”
The study investigated how ChatGPT and Bard categorize investor risk profiles compared to financial advisors. While there were no significant differences in the risk scores assigned by the chatbots and bankers for half of the clients, the chatbots often missed clients' specific characteristics. Chatbots may serve as a second opinion but cannot replace bankers.
The paper suggests that companies developing high-risk AI systems should demonstrate their safety before deployment, arguing for proactive risk management. It proposes a risk management approach where developers must provide evidence that risks are below acceptable thresholds. The paper discusses technical and operational evidence for safety, comparing its approach to the NIST AI Risk Management Framework.
The article explores the importance of critical infrastructure (CI) and essential services (ES) for population security and business continuity. It examines the challenges posed by the interdependence of CI and ES, which complicates threat identification and risk management. The study identifies new research directions on operational risk management, public security, and resilience in critical supply networks.