This paper examines the systematic legal transplantation phenomenon of the European Union's Artificial Intelligence Act (EU AI Act) among some Latin American countries, such as Brazil and Peru, through the lens of function-oriented comparison, taking anti-discrimination rules as an example. It has been discovered that, while these countries have generally introduced the EU's leading, risk-oriented regulatory model for new AI applications and safeguards against core interests, the resulting regulatory mismatch is substantial. Analysis finds deficiencies in this transplantation include the lack of corresponding institutional enforcement capabilities and inflexible risk classification systems that are not tailored to local priorities or technological situations. Although the papers deal with apparently identical themes such as anti-discrimination, in fact they have different socio-cultural and historical backgrounds. For example, in terms of anti-discrimination, although the focus of the EU is generally on problems such as immigration and religious discrimination, such as Islamophobia; The deep-seated reasons for discrimination in Latin America stem from colonial rule, impacting indigenous people. The paper has found that effective AI governance requires deep localisation modifications of transplanted regulatory models, and rules and implementation systems suitable for the specific cultural environment need to be developed; Only then can an appropriate regulation for the Digital Age be generated.
Ximing Yang (Fri,) studied this question.