Research on Algorithmic Ethics in Artificial Intelligence

With the rapid development of artificial intelligence, the ethical issues of artificial intelligence have become increasingly prominent. This paper discusses the problems of data security, bias and the implantation of morality and values in the ethics of artificial intelligence algorithms, and systematically and comprehensively expounds the main solutions to these three problems. It is found that the current three major technical paths for solving the data security issues have the problems of high dependency, large computation and communication overhead, and limited applicability, and the future trend is the integration and development of the three paths; the solution of the bias problem needs to be improved in terms of interpretability under the condition that the underlying fairness is difficult to be determined; and it is difficult for the implantation of morality and values to take into account the learning ability, adaptability, and interpretability at the same time. Finally, this paper analyzes and looks forward to the development of AI ethics, hoping to provide lessons and references for AI ethics-related research.