Nowadays, innovative systems incorporate machine learning processes to shape important decisions requiring absolute fair treatment for all users to mimic human motives. Cognitive biases function as systematic patterns of irrational decision-making in human judgment, which can unintentionally appear within AI systems and result from biased data accumulation, algorithmic creation, and human supervision. The research investigates AI system bias interactions with analysis of software development processes, which increases social inequities throughout the development cycle. The analysis includes three typical cognitive biases that affect AI model development: confirmation bias, anchoring Bias, and automation bias. The research seeks to reduce AI system bias, including utilizing varied data, performing bias audits, and deploying algorithms that focus on fairness. This document works to instruct developers and stakeholders about building equitable, trustable AI systems through biased education.