TIM-6072 v1

Polemi N, Praça I, Kioskli K and Bécue A (2024) Challenges and efforts in managing AI trustworthiness risks: a state of knowledge. Front. Big Data 7:1381163. doi: 10.3389/fdata.2024.1381163

This study explores gaps in AI risk management frameworks, particularly the lack of attention to human factors and the absence of metrics for socially related threats. The authors explain dimensions of trustworthiness, including legislation, AI cyber threat intelligence, and adversarial characteristics. The paper addresses socio-psychological threats including bias, misinformation, and privacy concerns. The paper calls for interdisciplinary collaboration, awareness campaigns, and continuous research efforts to build resilient AI aligned to ethical standards and societal expectations.

Ferrara, E. (2024). Fairness and Bias in Artificial Intelligence: A Brief Survey of Sources, Impacts, and Mitigation Strategies. Sci, 6(1), 3.

This paper provides an overview of fairness and bias in AI by exploring sources, societal impacts, and mitigation strategies. AI bias in critical areas such as healthcare, employment, criminal justice, and credit scoring, emphasize the influence of AI models shaping public perception. The article discusses the negative consequences of biased AI systems, including the reinforcement of inequalities and harmful stereotypes. The author underscores the importance of interdisciplinary collaboration, transparency, and accountability.

Ma, J. (2024). The impact of AI bias on social justice: challenges and solutions. Journal of Computing and Electronic Information Management, 15(3), 75-78. https://doi.org/10.54097/wz9v0f43

This author explores the impact of AI bias on social justice, particularly in critical areas such as healthcare, education, and employment. The study examines root causes of AI bias, including insufficient data diversity, lack of algorithmic transparency, and systemic biases in training datasets. The article emphasizes the urgency of addressing issues to ensure AI promotes fairness rather than discrimination.