Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{198366,
        author = {Isha Rajesh Tantak and Shraddha Mahendra Vetal and Prof.M.E.Maniyar},
        title = {Comparative Analysis of Bias in Predictive and Generative AI Systems},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {11569-11572},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=198366},
        abstract = {In recent years, there has been a swift incorporation of Artificial Intelligence (AI) in decision-making and content creation. In this research, an in-depth comparison is conducted to evaluate the level of biases in the field of predictive and generative AI. Predictive AI models such as Logistic Regression, Random Forest, and XGBoost were tested against fairness metrics on the Adult Income and COMPAS data sets. These include Statistical Parity Difference (SPD), Equal Opportunity Difference (EOD), and Disparate Impact (DI). To conduct the cross-paradigm comparison, the research employs a newly proposed metric called Unified Bias Score (UBS). The latter is a measure that combines the values of fairness metrics obtained during the prediction AI model testing and bias levels of the generative AI. The latter bias was determined using a structure that consists of 120 prompts from various sociocultural categories. From experimental results, it has been found that although XGBoost offers the maximum predictive performance of 0.8676, it also shows the maximum level of bias with regard to the fairness measures. The generative model shows significant explicit bias, where the percentage of gender bias is about 65%. The results reveal an important point of difference where predictive systems use implicit bias in decision surfaces while generative models use explicit bias in outputs.},
        keywords = {Algorithmic Bias, Artificial Intelligence, Ethical AI, Fairness Metrics, Generative AI, Machine Learning, Predictive AI, Unified Bias Score.},
        month = {April},
        }