Artificial Intelligence Biases in the Double Employment Discrimination of Women on Religious Grounds

Abstract

The use of artificial intelligence (AI) systems is transforming the labor market; however, their implementation without proper oversight can reinforce structural inequalities. Among vulnerable groups are women with visible religious convictions, who experience double marginalization based on both gender and religious affiliation. This discrimination manifests at three key stages: recruitment, the development of the employment relationship, and dismissal. In recruitment processes, hiring algorithms learn from historical data containing pre-existing biases, which can perpetuate exclusion. In addition, screening systems may penalize names associated with religious minorities or identify religious attire as an atypical trait. In Western cultures, the hijab, long skirts, the kippah, or certain visible Christian symbols may trigger cultural prejudices that are then transferred to the algorithm. Throughout the development of the employment relationship, AI systems used to assess performance, promotions, training selection, or occupational risks may entrench discriminatory barriers. These systems may even recommend the termination of employment for female workers whose religious practices require adjustments to working hours or dress codes, thereby reinforcing situations of exclusion. Human oversight, algorithmic audits, and transparency are crucial to building environments in which technology acts to guarantee inclusion rather than to perpetuate bias.