The expanding adoption of artificial intelligence systems across high-impact sectors has catalyzed concerns regarding inherent biases and discrimination, leading to calls for greater transparency and accountability. Algorithm auditing has emerged as a pivotal method to assess fairness and mitigate risks in applied machine learning models. This systematic literature review comprehensively analyzes contemporary techniques for auditing the biases of black-box AI systems beyond traditional software testing approaches. An extensive search across technology, law, and social sciences publications identified 22 recent studies exemplifying innovations in quantitative benchmarking, model inspections, adversarial evaluations, and participatory engagements situated in applied contexts like clinical predictions, lending decisions, and employment screenings. A rigorous analytical lens spotlighted considerable limitations in current approaches, including predominant technical orientations divorced from lived realities, lack of transparent value deliberations, overwhelming reliance on one-shot assessments, scarce participation of affected communities, and limited corrective actions instituted in response to audits. At the same time, directions like subsidiarity analyses, human-cent
The crypto space offers numerous opportunities for users to grow their wealth through trading, lending, and borrowing activities. However, these opportunities come with inherent risks that need to be carefully managed to protect your assets and maximize returns. By understanding the risks associated with wallets and depository services, trading, lending, and borrowing, users can make informed decisions and enjoy the benefits of the rapidly evolving world of cryptocurrencies. This review paper analyses 43 papers for the period of 2019–2023 and proposes recommendations for policy makers. The results confirm that international regulators expect national authorities to implement a regulatory framework for digital assets comparable to those that already exist for traditional finance. For national authorities, this means having and using the powers, tools and resources to regulate and oversee a growing market. Authorities should cooperate and coordinate with each other, at the national and international levels, to encourage consistency and knowledge sharing. Market operators (exchanges), service providers, exchanges and wallets, create effective risk management structures, as well as reliable mechanisms for collecting, storing, protecting and reporting data.
Public-Private Partnerships (PPPs) are mostly presented as a means to introduce efficient procurement methods and better value for money to taxpayers. However, the complexity of the PPP mechanism, their lack of transparency, accounting rules and implicit liabilities make it often impossible to perceive the amount of public expenditure involved and the long-run impact on taxpayers, providing room for fiscal illusion, i.e., the illusion that PPPs are much less expensive than traditional public investments. This psaper, thanks to a systematic review of the literature on the EU countries experience, tries to unveil the sources of this illusion by looking at the reasons behind the PPPs’ choice, their real costs, and the sources of fiscal risks. The literature suggests that PPPs are more costly than public funding, especially when contingent liabilities are not taken into account, and are employed as mechanisms to circumvent budgetary restrictions and to spend off-balance. The paper concludes that the public sector should share more risks with private sectors by reducing the amount of guarantees, and should prevent governments from operating through a sleight of hand that deflects attention away from off-balance financing, by applying a neutral fiscal recording system.
Resisting the adoption of medical artificial intelligence (AI), it is suggested that this opposition can be overcome by combining AI awareness, AI risks, and responsibility displacement. Through effective integration of public AI dangers and displacement of responsibility, some of these major concerns can be alleviated. The United Kingdom’s National Health Service has adopted the use of chatbots to provide medical advice, whereas heart disease diagnoses can be made by IBM’s Watson. This has the ability to improve healthcare by increasing accuracy, efficiency, and patient outcomes. The resistance may be due to concerns about losing jobs, anxieties about misdiagnosis or medical mistakes, and the consciousness of AI systems drifting more responsibility away from medical professionals. There is hesitancy among healthcare professionals and the general public about the deployment of AI, despite the fact that healthcare is being revolutionised by AI, its uses are pervasive. Participants’ awareness of AI in healthcare, AI risk, resistance to AI, responsibility displacement and ethical considerations were gathered through questionnaires. Descriptive statistics, chi-square tests and correlation analyses were used to establish the relationship between resistance and medical AI. The study’s objective seeks to collect data on primary and public AI awareness, perceptions of risk and feelings of displacement that the professionals have regarding medical AI. Some of these concerns can be resolved when AI awareness is effectively integrated and patients, healthcare providers, as well as the general public are well informed about AI’s potential advantages. Trust is built when, AI related issues such as bias, transparency, and data privacy are critically addressed. Another objective is to develop a seamless integration of risk management, communication and awareness of AI. Lastly to assess how this comprehensive approach has affected hospital settings’ ambitions to use medical AI. Fusing AI awareness, risk management, and effective communication can be used as a comprehensive strategy to address and promote the application of medical AI in hospital settings. An argument made by Chen et al. is that providing training in AI can improve adoption intentions while lowering complexity through the awareness of AI.
The state delivery of affordable and sustainable housing continues to be a complicated challenge in Africa, and there is a need to encourage private sector participation. As a result, this study examines the risks associated with private sector participation in affordable housing and supporting infrastructure investment and the strategies towards mitigating the risks from an Afrocentric perspective. The evidence from a systematic literature review was coupled with the opinion of an international expert panel to address the paper’s aim and provide recommendations for developing improved housing and supporting infrastructure in Sub-Saharan Africa. The review outcomes and the qualitative data from the panel discussion were analysed using thematic analysis. The results revealed that market dynamics, land supply and acquisition constraints, cost of construction materials, unsupportive policies, and technical and financial factors constitute risks to affordable housing in the region. Mitigation strategies include leveraging joint efforts, strengths, and resource bases, increasing access to land and finance for private sector participation, developing a supportive government framework to promote an enabling environment for easy access to land acquisition and development finance, local production of building materials, research and technology adoption. In line with the United Nations (UN) Agenda 2030 targets and principles, reforms are required across the housing value chain, involving the private sector and community. Application of the study’s recommendations could minimise the risks of affordable housing delivery and enhance private sector participation.
This study introduces an innovative approach to assessing seismic risks and urban vulnerabilities in Nador, a coastal city in northeastern Morocco at the convergence of the African and Eurasian tectonic plates. By integrating advanced spatial datasets, including Landsat 8–9 OLI imagery, Digital Elevation Models (DEM), and seismic intensity metrics, the research develops a robust urban vulnerability index model. This model incorporates urban land cover dynamics, topography, and seismic activity to identify high-risk zones. The application of Landsat 8–9 OLI data enables precise monitoring of urban expansion and environmental changes, while DEM analysis reveals critical topographical factors, such as slope instability, contributing to landslide susceptibility. Seismic intensity metrics further enhance the model by quantifying earthquake risk based on historical event frequency and magnitude. The calculation based on higher density in urban areas, allowing for a more accurate representation of seismic vulnerability in densely populated areas. The modeling of seismic intensity reveals that the most susceptible impact area is located in the southern part of Nador, where approximately 50% of the urban surface covering 1780.5 hectares is at significant risk of earthquake disaster due to vulnerable geological formations, such as unconsolidated sediments. While the findings provide valuable insights into urban vulnerabilities, some uncertainties remain, particularly due to the reliance on historical seismic data and the resolution of spatial datasets, which may limit the precision of risk estimations in less densely populated areas. Additionally, future urban expansion and environmental changes could alter vulnerability patterns, underscoring the need for continuous monitoring and model refinement. Nonetheless, this research offers actionable recommendations for local policymakers to enhance urban planning, enforce earthquake-resistant building codes, and establish early warning systems. The methodology also contributes to the global discourse on urban resilience in seismically active regions, offering a transferable framework for assessing vulnerability in other coastal cities with similar tectonic risks.
Copyright © by EnPress Publisher. All rights reserved.