Data Quality in the Age of AI: A Review of Governance, Ethics, and the FAIR Principles

Data quality is fundamental to scientific integrity, reproducibility, and evidence-based decision-making. Nevertheless, many datasets lack transparency in their collection and curation, undermining trust and reusability across research domains. This narrative review synthesizes scientific and technical literature published between 1996 and 2025, complemented by international standards (ISO/IEC 25012, ISO 8000), to provide an integrated overview of data quality frameworks, governance, and ethical considerations in the era of Artificial Intelligence (AI). Sources were retrieved from PubMed, Scopus, Web of Science, and grey literature. Across sectors, accuracy, completeness, consistency, timeliness, and accessibility consistently emerged as universal quality dimensions. Evidence from healthcare, business, and public administration suggests that poor data quality leads to substantial financial losses, operational inefficiencies, and erosion of trust. Emerging frameworks are increasingly integrating FAIR principles (Findability, Accessibility, Interoperability, Reusability) and incorporating ethical safeguards, including bias mitigation in AI systems. Data quality is not solely a technical issue but a socio-organizational challenge that requires robust governance and continuous assurance throughout the data lifecycle. Embedding quality and ethical governance into data management practices is crucial for producing trustworthy, reusable, and reproducible data that supports sound science and informed decision-making.