Engineering equity: designing diversity-aware AI to reflect humanity

Engineering equity: designing diversity-aware AI to reflect humanity

연구 분야: Artificial Intelligence

학회: AI & SOCIETY

Diversity plays a crucial role in recommendation systems. Enhancing the diversity of recommendations can expand users’ perspectives, improve user experience, and support social equity. Developing diversity-aware AI is essential for creating systems that are adaptive, ethical, and capable of reflecting the complexity of human society. The necessity to create diversity-aware AI stems from the understanding that if AI is to mimic human intelligence in meaningful ways, it must surpass static, monolithic models that narrowly reflect only a portion of the human experience. AI must embrace diverse perspectives, adapting not only to the varying needs and backgrounds of users but also to changes in societal understanding. Creating more human-like AI requires focusing on the diverse reasoning and behavior of artificial agents and developing systems capable of dealing with such diversity is key to achieving more human-like AI. This study discusses the necessity of diversity in AI, arguing that it is essential for overcoming the limitations of static models, incrementally combining different components of intelligence, and expanding the notion of what constitutes intelligent adaptation.