As a scientific research field that attempts to study, imitate and expand human intelligence, artificial intelligence has been accompanied by profound technical ethical debates since its birth. With the breakthrough progress and rapid implementation of machine learning and other related work in recent years, ethical issues have become increasingly prominent and forced the academic community and society to begin to face the ethical governance challenges of this technology. Although initial progress has been made in the normative research on ethical governance, its governance practices are still difficult to implement, and ethical practices show a trend of gradually lagging behind the needs of technological development. Therefore, establishing an ethical governance practice plan that matches the ever-developed artificial intelligence technology and realizing a benign interaction between governance theory and governance practice will be a key issue in the future development of the field of artificial intelligence. The abstraction of ethical governance theory has made it difficult to implement the ethical principles of artificial intelligence at present, and artificial intelligence ethical computing (AI) will be an important solution to this challenge. This study clarifies the importance of ethical computing by exploring the necessity of reality and the possibility of development. Based on relevant research, the research scope of ethical computing is given. According to the calculation process, the degree of cognition of ethical mechanism and the degree of autonomy of system ethical decision-making is divided. Two types of research paradigms of high-order cognition and low-order cognition of ethical computing are established. Three levels of calculation of ethical measurement, ethical decision-making and ethical reasoning are abstracted according to their calculation stage. This ethical computing framework can sort out the current application of ethical computing. This article uses ethical embedding and fair machine learning as examples to illustrate the research characteristics and technical methods of the two types of research paradigms. On this basis, the construction of an ethical governance system with ethical computing as the core is further discussed and constructed, and analyses the possible solutions to resolve ethical governance difficulties through ethical computing, and makes a prospect for the development of artificial intelligence ethical computing.

The improvements in massive data, basic computing power, intelligent algorithms, special hardware and other aspects have prompted the booming development of the artificial intelligence industry, and various technologies have continuously reshape human living habits and production methods. The application breadth of related technologies is constantly expanding and the depth is constantly deepening, which has given birth to various cross-fusion fields including AI for [1], quantum machine learning research [2], etc. New technologies are emerging one after another and new applications are being implemented, bringing unprecedented opportunities to the development of the field of artificial intelligence. However, at the same time, technical ethical anxiety is increasing, and effective ethical governance of technology is imminent. What ethical issues are facing the current development of artificial intelligence technology? What are the difficulties in ethical governance? What technical means can assist ethical governance? This article proposes to build a technical tool system for ethical governance with artificial intelligence ethical computing (AI, referred to as ethical computing) technology as the core. And what is ethical calculation? What techniques are used to perform ethical calculations? How can we build an ethical governance system through this technical means? The discussion will be launched later. Starting from the ethical issues we face, we will start with the following three typical application scenarios as examples. (1) Artificial intelligence technology has brought major changes to the medical and health field, mainly reflected in two forms of technical support: virtual branch and physical branch. Virtual branch refers to artificial intelligence algorithms that can use big data to mine potential medical auxiliary information, including protein response prediction, drug prediction, psychological rehabilitation auxiliary treatment, etc.

The physical branch refers to medical service robots supported by various artificial intelligence algorithms, including robot partners used to care for critically ill patients, assistant doctors and even surgeons in surgical operations [4]. In these important application scenarios that are related to human life and health, ethical issues are particularly prominent. It involves the issue of how to protect the privacy of the stakeholders from being violated [5], and also involves the difficulty of ensuring that decisions do not cause damage to the life and health of the audience. For example, in the application scenario of artificial intelligence surgeons, how should we define whether the occurrence of an surgical accident is an accident? Who is responsible for the accident? Such problems involve various complex ethical subjects, and the ethical issues behind them are often very complex. Is it possible to help define these complex ethical issues through technical methods? (2) Self-driving cars are expected to improve transportation efficiency and reduce the probability of traffic accidents [6]. However, due to its high degree of autonomy, this scenario also involves very complex ethical issues: how do autonomous driving machines make moral decisions and how do they determine the responsibility for the accident? How should society quantify the guiding ethical principles of machine behavior? This has become an urgent question to be responded to. Related research [7] conducted an extensive social experiment on major challenges in this field, collected 40 million ethical decisions from 233 countries and regions and conducted statistical analysis, pointing out that moral trade-offs show a trend of unity from a macro perspective, but there are still internal conflicts, interpersonal differences, regional and cultural differences in moral decision-making.

To reduce conflicts, this field also requires more ethical dialogue and more precise technical quantitative standards. (3) Computer-aided decision-making is also one of the important applications of artificial intelligence. The effective mining of massive data by algorithms allows them to deeply explore historical decisions and learn decision-making elements from them. The improvement in the automation of decision-making methods can greatly improve decision-making efficiency, but the problems of bias and discrimination hidden behind it are also worrying. Typical scenarios include systems for estimating the probability of recidivism, automatic screening of recruitment resumes, advertising push, etc. [8]. These artificial intelligence used for decision-making will have a significant impact on society and individuals, and the transparency and fairness of its decision-making mechanism need to be improved urgently. Not only decision-making systems, but also in various studies such as natural language processing [9] and computer vision [10, 11], there are problems of learning and amplifying historical biases. How to improve these problems still needs more exploration. At the same time, the relevant proposals for auxiliary decision-making scenarios also pointed out [12] that the application of artificial intelligence in decision-making may lead to changes in organizational culture and individual behavior, so it is also necessary to develop practical evolutionary indicators of the impact of artificial intelligence technology to measure its benefits and its long-term impact on decision-making stakeholders. The above are ethical problems existing in some classic application scenarios. The application of new technologies is constantly raising new challenges to ethical research. For example, AI painting and generated large models represented by AI have recently triggered discussions on copyright disputes and plagiarism issues [13]. On the basis of exploring the progress of ethical computing, the problems caused by such large models will also be briefly discussed.

It is not difficult to see that current artificial intelligence technology has caused a wide range of profound social problems, and with the development of technology, it will face more complex technical ethical dilemmas, so it is urgent to provide effective response solutions. In fact, the discussion on the ethical issues of artificial intelligence [14,15] was born as early as 1960, and the research on the ethical research of artificial intelligence accompanied the development of technology. However, early research mainly revolved around ethical theory, separated from specific application scenarios and was highly abstract. Therefore, how to effectively consider ethical factors in actual application scenarios and build a practical and feasible ethical governance plan is an important research topic. In response to ethical governance issues, in recent years, various national organizations have made important efforts, put forward many technical ethical demands including explainability, fairness, privacy, etc., and formulated relevant industry development standards. Despite many legal norms and initiatives, due to the ambiguity and differences of these abstract indicators, the practice of ethical governance still faces many difficulties. This article believes that concreteizing ethical norms, quantifying ethical appeals and providing technical support for ethical norms is the key to breaking the deadlock, that is, ethical calculations eliminate abstract ambiguity and provide quantitative characteristics, which is expected to make up for the differences between theory and practice. However, the key problem facing ethical calculation is: Can abstract ethical concepts be calculated? How to calculate? How should computing serve ethical governance practices? This article intends to start from the current development status and difficulties of artificial intelligence ethics research, and points out the necessity of research on ethical computing. By exploring the research and development of ethical computing, discussing the possibility of ethical computing and defining the concept of ethical computing. Further addressing the problem of how to calculate, we summarize the existing calculation methods and calculation paradigms of ethical calculation, and give examples of representative ethical calculation techniques. Finally, the relationship between calculation and ethical governance will be analyzed.

The overall discussion idea is shown in Figure 1, and the explanation of ethical computing from the computing system to the ethical governance system is completed. Specifically, the second section summarizes and explores the practical background and computing history of ethical computing research. It explains the current development of ethical research, points out the development pain points of fuzzy ethical theory and limited practical feasibility in ethical governance, points out the important practical significance of ethical computing, and analyzes the research background of computational ethics on this basis. The third section gives the specific definition of ethical computing in this article, summarizes the research goals, research methods and research paradigms of ethical computing, and gives two representative work, and uses fair machine learning as a case to give examples of ethical computing. Section 4 constructs an ethical governance system based on ethical computing, and finally, in Section 5, ethical computing and ethical governance are outlook.

Convenient and easy to view

Convenient download, please follow the official account of "Talk" (click the blue "Talk" above to follow)

Knowledge, professional and trustworthy artificial intelligence knowledge distribution, making cognitive collaboration faster and better! Welcome to register and log in to obtain knowledge and information on the topic of AI (AI and military, medicine, public security, etc.)!

Welcome to scan on WeChat to join the expert artificial intelligence knowledge planet group to obtain the latest AI professional knowledge tutorial materials and communicate with experts!

Click "Read original text" to learn about usage knowledge, view and obtain AI topic knowledge materials