Artificial Intelligence (AI) has transformed systems that support data-driven decision-making processes and added new functionalities and capabilities in different sectors, such as fisheries and health. Data bias in AI systems can produce results that influence decision-making and could even cause already-existing disparities. The ripple effect of data bias on AI-enabled decision-making was examined in this study. This study was conducted by using the IBM 360 AI Fairness toolkit (AIF 360) and traditional machine learning models to analyse Diabetes and Fisheries data to know the bias, skewness, and risk factors. Our research presented empirical findings that emphasise the need for diverse and representative datasets to use AI decision-making to ensure equitable and sustainable outcomes and to avoid inequalities and injustices in the results. The results revealed that the accuracy of datasets increased from 82 to 89% and from 51 to 72% after using the reweighted algorithm of the AIF 360 tool, which balanced the bias to avoid misdiagnosis and misallocation of resources. This research was conducted on the relevance of obtaining balanced and constituency-balanced data to create AI systems that would be able to make effective decisions with fairness whilst serving the needs of diverse groups. Future research should focus on developing tools to mitigate bias while improving decision-making.