Explore publications.

The integration of artificial intelligence (AI) into human resource management (HRM) processes has significantly enhanced organisational operations by optimising efficiency in talent acquisition, performance assessment, and employee engagement initiatives. Nonetheless, this integration introduces ethical challenges, including algorithmic bias, data privacy concerns, and the reduction of human decision-making autonomy. The purpose of the review is to systematically examine the ethical considerations associated with AI deployment in HRM, explicitly addressing issues related to bias, transparency, privacy breaches, and job displacement. Additionally, it evaluates strategies for deploying AI responsibly to uphold principles of fairness, accountability, and stakeholder trust. Considering the purpose of the review, a comprehensive systematic review of peer-reviewed literature was conducted, sourcing data from Scopus, Web of Science, and Google Scholar. The review targeted publications relevant to the aims of the study, utilising search keywords such as "ethical consideration," "algorithmic bias," and "data privacy." Data synthesis was conducted via narrative methods complemented by thematic analysis to identify prevailing patterns, research gaps, and emergent themes. Findings revealed that AI enhances HRM functions through automation, predictive analytics, and the delivery of individualised employee experiences, contributing to cost reductions and improved decision-making efficacy. However, potential negative implications include reinforcement of existing biases, violations of privacy, lack of transparency in decision-making processes, and threats to employment stability. Empirical evidence indicates disproportionate impacts on marginalised populations and underscores the necessity for ethical frameworks, including the development and implementation of explainable AI (XAI) and hybrid human-AI systems. Finally, it is affirmed that although AI offers transformative operational benefits within HRM, the associated ethical risks necessitate the implementation of comprehensive governance structures, bias mitigation protocols, and active stakeholder participation. Approaches such as hybrid human-AI systems, targeted workforce training, and adherence to regulatory standards are essential to ensuring sustainable and ethical AI integration. Future research should prioritise investigating employee perceptions and conducting cross-cultural analyses to enhance ethical AI governance in HRM contexts.

In the context of rapid technological advancement and evolving financial ecosystems, this study investigates the interplay between financial literacy, digital finance, and financial behaviour among high school teachers in KwaZulu-Natal (KZN), South Africa. Structural equation modelling (SEM) approach was used to describe intricate relationships between latent constructs, namely, financial literacy, financial Behaviour and, digital finance Knowledge as well as observed variables (B7 to B22, C23 to C44, D45 to D56, demographics). Demographic variables include Gender, Age Category, Qualification Type, Teaching Duration, Financial Literacy Course and Field Employed. The model defined each construct as a linear combination of its observed variables as well as demographic variables under its measurement component, and hypothetical relationships among three latent variables under its structural component. We used Python software and machine learning code to create and estimate the model using observed data. The missing values in each observed variable were imputed with averages of the variables’ data. The tests of normality assumption for SEM were performed using histogram and q-q plots, Shapiro-Wilk test D'Agostino and Pearson's test.  Most data vectors violated the normality assumption. Diagonally Weighted Least Squares (DWLS) was used for parameter estimation as a suitable method that can provide better estimation for more complex models with many parameters. This method has an additional advantage of being less sensitive to outliers. The DWLS parameter estimation results, based on p-values, suggest that financial literacy and financial behaviour have a negative association which means that lacking financial behaviour results in financial literacy. A positive association between has been identified between financial behaviour and digital finance knowledge. Financial literacy and digital finance knowledge have a weak or no relationship. Thus, having or not having digital finance knowledge does not necessarily determine the possibility of experiencing financial literacy. Furthermore, among demographic variables, the financial literacy Course is the only demographic variable that has a significant impact, and only on Financial Behaviour and Financial literacy. The rest of the variables do not significantly influence the latent constructs. This study uniquely applies SEM using the DWLS estimator to explore personal finance dynamics in an emerging economy, offering concise insights into key relationships.