人工智能教育的算法风险与善治

Algorithmic Risk and Good Governance of Artificial Intelligence Education

迈入人工智能时代,人工智能教育已是教育发展的大势所趋,而算法是人工智能教育的强中之基。算法在为人工智能教育带来快速发展的同时,也带来了不可预知的风险:算法的既定性导致教育被浅层化,“黑箱”效应催生了教育被隐形控制、算法偏见加剧了教育不公,以及算法的大规模应用易引发教育同质化,其危害不容小觑。深入究因发现,算法的固有属性与教育的特质之间的博弈是“风险”产生的本体的、深层次的缘起。防范人工智能教育应用的算法风险,让算法遵循“善法”,可通过优化算法设计、增加算法的透明度、减少算法偏见滋生、确立算法问责机制,稳步推进人工智能教育算法从善的深化发展。

Entering the era of artificial intelligence,artificial intelligence education has become the general trend of education development,and algorithms are the strong foundation of artificial intelligence education.While the algorithm brings about the rapid development of AI education,it also brings unpredictable risks,and these dangers cannot be underestimated,these risks are:the established characteristics of the algorithm leads to the shallow of education,the“black box”effects give birth to the invisible control of education,and the algorithm bias exacerbating inequities in education,and the large-scale application of algorithms that have contributed to the convergence of education.In-depth research found that the game between the inherent attributes of the algorithm and the characteristics of education is the ontological and deep origin reasons of the algorithmic risk.To prevent and control the algorithmic risks of artificial intelligence education,so that algorithms follow the“good law”,we can optimize the algorithm design,increase the transparency of the algorithm,reduce the breeding of algorithm bias,establish an algorithm accountability mechanism,and steadily promote the deepening development of artificial intelligence education algorithms.