Recently, the European Commission issued the ethical guidelines for artificial intelligence, which lists seven key conditions for "trusted artificial intelligence" - human initiative and supervision capabilities, security, privacy data management, transparency, inclusiveness, social welfare, Accountability mechanism to ensure that artificial intelligence is safe and reliable enough.

In fact, although people have different opinions on where artificial intelligence will go in the future, it has become a basic consensus to impose ethical regulation of artificial intelligence.

"In the era of artificial intelligence, big data and 5G, it is necessary to make advanced technologies better serve people and make norms." Cui Dongshu, Secretary-General of the National Passenger Car Market Information Joint Conference Talk about it. This is a "peace of mind" for technological research and development based on artificial intelligence or the development of the industry.

Countries around the world actively launch ethical rules of artificial intelligence

June 2016

Ethics Committee Member of the Japanese Artificial Intelligence Society

A draft ethical outline for researchers was drafted. The draft proposes the possibility that AI brings harm to human society and requires measures to be taken to eliminate threats and prevent malicious use.

September 2016

British Standard Industry Association

The industry's first public indicator on robot ethical design, namely the "Guidelines for Ethical Design and Application of Robots and Machine Systems", aims to ensure that intelligent robots produced by humans can be integrated into the moral norms of human society.

September 2016

Artificial Intelligence Cooperation Organization

Five major tech giants including Amazon, Google, Microsoft and IBM have announced the establishment of a non-profit organization. One of the goals is to provide a demonstration of the research on artificial intelligence, involving areas such as ethics, equity and inclusion.

November 2016

American Institute of Electrical and Electronic Engineers (IEEE)

Release the world's first "Draft Design of the Ethical Code of Artificial Intelligence ()". I want to use this document to help the technology industry build AI automation systems that benefit humans and change moral ethics without worrying.

January 2017

Al Conference in Asiloma, California

Nearly a thousand experts in the field of artificial intelligence related to artificial intelligence jointly signed the famous "23 Principles of Asiloma Artificial Intelligence".

November 2017

American Institute of Electrical and Electronic Engineers (IEEE)

Three new AI ethical standards have been announced: ethical driving standards for machined systems, intelligent systems and automatic systems, fail-safe design standards for automatic and semi-automatic systems, and ethical AI and measurement of well-being of automated systems.

April 2018

European Commission

The European Union Artificial Intelligence document was released to propose the EU path of artificial intelligence. Among them, establishing an appropriate moral and legal framework is one of the three-pronged approaches to the development of artificial intelligence.

April 2018

Special Committee on Artificial Intelligence in the House of Lords

The release of the report said that in the process of developing and applying artificial intelligence, it is necessary to put ethics at the core to ensure that this technology better benefits humanity.

April 2019

European Commission

The official version of "For AI" was released, which proposed a framework for realizing the entire life cycle of trustworthy artificial intelligence.

Information comes from global technology map

What ethical challenges will artificial intelligence bring

In the report "The Foundation of Future - Social Roles and Ethics of Artificial Intelligence" released by the China Development Research Foundation, it is proposed that from a humanistic perspective, artificial intelligence has brought some of the things that are already happening or about to happen and may shake our social foundation. fundamental problem.

With the continuous development of artificial intelligence, our understanding of humans is becoming more and more materialized and desensible, and the boundaries between humans and machines are becoming more and more blurred. We need to think about the consequences of this blurred boundary. How should we treat the relationship between machines and ourselves? Should humans and machines be integrated? If people's feelings about happiness and pain can be met through other physical and chemical means, then do people still have the need and motivation to participate in social interaction?

Artificial intelligence also brings new social power structure problems. With artificial intelligence, enterprises can give each user a large amount of data labels and understand people's preferences and behaviors based on these labels, even exceeding users' knowledge of themselves. This is a huge right asymmetry.

Furthermore, artificial intelligence may cause bias reinforcement. In social media, artificial intelligence recommends people with similar opinions to each other, and news push often has path dependence. As people's sources of information become increasingly dependent on intelligent machines, biases are reinforced in this assimilation and path dependence.

Artificial intelligence also greatly amplifies the information and knowledge processing capabilities of society, and the redundancy of information and knowledge actually makes people fall into the dilemma of choice. If the number and scope of people's participation in social interactions are narrower, and artificial intelligence is increasingly involved in the production of knowledge, the relationship between knowledge and human needs will become more and more indirect, and will even dominate human needs in turn.

Although artificial intelligence will also promote further specialized division of labor and create new job opportunities, not everyone has the ability to overcome technical and social barriers. Yuval Harari warned in a brief history of the future that "useless classes" may appear in the future society, and this concern is not completely unreasonable.

In addition, artificial intelligence also challenges traditional laws, regulations and social norms. For example, if an accident occurs in a driverless car, should we attribute it to the company that develops the product, the product owner, or the artificial intelligence product itself?

How to build ethical norms for artificial intelligence

At present, some leading artificial intelligence companies are already paying attention to the ethical issues of artificial intelligence development and application.

Microsoft CEO Schbert said that to design trustworthy artificial intelligence, solutions that reflect ethical principles must be adopted. Therefore, Microsoft proposed six basic ethical guidelines: fairness, inclusiveness, transparency, responsibility, reliability and security, privacy and confidentiality. .

Google also stated that in artificial intelligence development applications, it adheres to seven guidelines including fairness, security, transparency, and privacy protection, and clearly lists "AI applications that Google will not pursue."

The report believes that the forward-looking exploration of these companies is very important. However, considering the far-reaching impact of artificial intelligence on the future society, it is also necessary to work together by the whole society and governments of all countries to formulate ethical norms and policy directions for the development and application of artificial intelligence, laying the foundation for its healthy future development.

To this end, the report made a series of suggestions—

Social level: Universities and research institutions carry out forward-looking scientific and technological ethics research to provide theoretical support for the establishment of relevant norms and systems; artificial intelligence enterprises should include ethical considerations in the corporate social responsibility framework; investment institutions should include ethical issues in the ESG (environment, Social and governance framework guides enterprises to carry out responsible artificial intelligence product development; social organizations can promote the construction of artificial intelligence ethical norms through training, publishing ethical evaluation reports, summarizing representative cases, etc.

Public policy level: policies for the research and development and application of artificial intelligence should put people at the core, meet the needs of people's comprehensive development, and promote the fair and sustainable development of society; the government needs to set up special funds to support universities and research institutions in the development of cutting-edge AI and other artificial intelligence. Ethical research of science and technology; prioritize the application of artificial intelligence to solve prominent challenges in the social field, including reducing poverty and inequality, promoting the integration of vulnerable groups into society and participating in the social development process; an artificial intelligence ethics committee composed of government departments and industry experts should be established , provide ethical guidance on the development and application of artificial intelligence, and conduct ethical and legality assessments on artificial intelligence products with significant public impact.

International level: Xue Lan, dean of the School of Public Administration of Tsinghua University, said, "The development of artificial intelligence will have a profound impact on the current international order in three aspects: innovative governance, sustainable development and global security cooperation, and requires governments and all sectors of society to go from humanity. The community with a shared future is of great concern and response. Only by strengthening cooperation and exchanges between countries can we truly build a new global, co-construction and sharing, safe, efficient and sustainable development order of artificial intelligence governance."

The report also recommends that governments should promote data sharing technology to provide shared public data sets for artificial intelligence training and testing. On the premise that personal information is protected, promote the free circulation of data, strengthen international cooperation, and establish a multi-level international artificial intelligence governance mechanism. Governments should incorporate the development of artificial intelligence into the international cooperation agenda through the United Nations, G20 and other international platforms, and use artificial intelligence to promote the realization of the United Nations 2030 Sustainable Development Goals.

▼

Previous recommendations (click to read)