Gender Bias and Cultural Misrepresentation in AI: A Critical Inquiry into Cross-Cultural Communication and Algorithmic Design

Against the backdrop of gender disparities in artificial intelligence datasets, this study examines their impact on intercultural communication, particularly with regard to elicitive processes and shame- and guilt-oriented conscience formations in the context of individualistic and collectivist cultures. In relation to this causal and exploratory context, historical biases in the underlying datasets become apparent in the development and implementation of AI-powered applications, including translation services, recruitment algorithms, and content moderation systems, thereby perpetuating gender stereotyping. Under these conditions, the Western conception of AI models, particularly those developed in the US and Europe, reveals significant shortcomings in interpreting non-Western communication and memorization patterns. This results in inadequate representations and the potential marginalization of users in digital interaction contexts, which contradicts the requirements of DIN EN ISO 9241. In this regard, the empirical findings of the study indicate an insufficient integration of heuristic complexity-reducing bias mitigation strategies and ethical AI frameworks in the development process of artificial intelligence. However, these findings also highlight the urgent need to establish algorithmic fairness standards. This requires an interdisciplinary discourse among computer scientists, UX designers, ethicists, didacticians, and cultural and communication scientists, fostering collaborative research practices in the spirit of a consortial research approach. As a result, future research projects should explore AI-time interactions, bias reduction methods, and marginalized cultural and gender-specific conceptualizations. The analysis should be guided by the standards of DIN EN ISO 9241-11 (usability), 9241-110 (principles of interaction), and 9241-210 (human-centered design of interactive systems). Additionally, both the European Accessibility Act (EAA) and the Barrierefreiheitsstärkungsgesetz (BFSG), which will take effect on June 28, 2025, impose corresponding regulatory implications. From this perspective, inclusive AI governance structures can be identified that contribute to the creation of equitable AI models, ultimately promoting the development of more target group-oriented and efficient digital communication modalities for global communities.