Woman in the Mirror of AI: How Is Gender Fairness Ensured in Artificial Intelligence Decisions?

Artificial intelligence (AI) is rapidly entering fields that directly affect human life, including medicine, recruitment, credit assessment, security, education, and public services. However, the fact that algorithmic decisions are automated does not mean that they are always impartial. On the contrary, AI may reproduce or amplify gender imbalances present in the data on which it is trained. Discussions at the 2026 ITU WSIS Forum highlighted the need for sex-disaggregated data and separate evaluation of outcomes for high-risk AI systems [1]. The International Labour Organization also notes that the impact of generative AI is not gender-neutral and that women are concentrated in occupations that are more susceptible to automation [2]. This article examines the problem of gender fairness in AI systems across the stages of data collection, algorithmic modelling, evaluation, and final decision-making. It also proposes a conceptual “4D Gender-Fair AI” model for developing gender-sensitive AI: Data, Design, Disaggregation, and Decision.