AI Governance for Dark Data: Ethical Management of Unlabeled and Unstructured Data in AI Systems

Shaik Abdul Kareem

Abstract: The increasing reliance on artificial intelligence (AI) systems in various sectors has led to an exponential growth in the amount of data being processed. However, a significant portion of this data remains "dark"?unlabeled, unstructured, and often ignored in traditional AI governance frameworks. This paper proposes a novel governance framework specifically designed to manage the ethical risks associated with dark data in AI systems. By focusing on fairness, transparency, and accountability, the proposed framework aims to ensure that dark data is utilized responsibly, mitigating potential biases and ensuring compliance with ethical standards. Through real-world applications in finance and healthcare, the paper demonstrates the practical impact of this governance model, providing empirical evidence of its effectiveness in improving data management practices and gaining industry recognition.

Keywords: AI Governance, Dark Data, Ethical AI, Unlabeled Data, Unstructured Data, Data Management

How to Cite?: Shaik Abdul Kareem, "AI Governance for Dark Data: Ethical Management of Unlabeled and Unstructured Data in AI Systems", Volume 11 Issue 9, September 2022, International Journal of Science and Research (IJSR), Pages: 1273-1278, https://www.ijsr.net/getabstract.php?paperid=SR22910213254, DOI: https://dx.doi.org/10.21275/SR22910213254