Scholarlink

<p>This paper introduces a novel framework for integrating cross-cultural perspectives into AI ethics governance using Cross-Cultural Transfer Learning (CCTL). This approach addresses the challenges posed by diverse cultural norms in global AI systems, particularly in healthcare, education, and business. We propose a methodology that combines transfer learning with ideological education to adapt AI systems to culturally specific ethical standards. Current AI ethics frameworks often rely on Western-centric norms, which fail to account for the cultural diversity in global AI governance. This can lead to biased or ethically misaligned systems in non-Western contexts. The CCTL Framework bridges this gap by integrating transfer learning with ideological education to align AI systems with local cultural norms. Evaluated on four datasets, including Global AI Ethics and Healthcare AI Ethics, CCTL achieves 90.5% accuracy, outperforming baselines in fairness (+12%) and transparency metrics. This work proposes a framework for improving cultural adaptability in AI ethics governance and demonstrates its potential across multiple domains.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Cross-cultural transfer learning framework for ideological education in global AI ethics governance

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月25日

查看全文
  • Jiaoyang Lv,
  • Yu Dong

摘要

This paper introduces a novel framework for integrating cross-cultural perspectives into AI ethics governance using Cross-Cultural Transfer Learning (CCTL). This approach addresses the challenges posed by diverse cultural norms in global AI systems, particularly in healthcare, education, and business. We propose a methodology that combines transfer learning with ideological education to adapt AI systems to culturally specific ethical standards. Current AI ethics frameworks often rely on Western-centric norms, which fail to account for the cultural diversity in global AI governance. This can lead to biased or ethically misaligned systems in non-Western contexts. The CCTL Framework bridges this gap by integrating transfer learning with ideological education to align AI systems with local cultural norms. Evaluated on four datasets, including Global AI Ethics and Healthcare AI Ethics, CCTL achieves 90.5% accuracy, outperforming baselines in fairness (+12%) and transparency metrics. This work proposes a framework for improving cultural adaptability in AI ethics governance and demonstrates its potential across multiple domains.