On August 29, at the 2024 China Internet Civilization Conference held in Chengdu,Generative AIIndustry self-regulatory initiatives》is officially released.
The "Generative Artificial Intelligence Industry Self-Discipline Initiative" issues industry initiatives from the aspects of ensuring the security and compliance of data and algorithm models, promoting the construction of content ecology, pursuing technological innovation and quality improvement, following values and ethical standards, and promoting exchanges, cooperation and open co-construction.
Representatives of the units participating in the initiative said that implementing the initiative is not only a solemn commitment to the general public, but also an inherent requirement for industry development and corporate operations.
The list of contributors to the "Generative Artificial Intelligence Industry Self-Discipline Initiative" includes well-known companies such as Alibaba Cloud, Baidu Netcom, and Tencent, as well as universities such as Peking University and Tsinghua University.
IT Home attaches the full text of the "Generative Artificial Intelligence Industry Self-Discipline Initiative":
- As science and technology develop rapidly, the field of generative artificial intelligence has shown unprecedented development momentum and potential. As this emerging technology progresses, we must deal with its risks and challenges. In order to ensure the healthy and sustainable development of generative artificial intelligence technology and industry and create greater value for society, we solemnly launch the self-discipline initiative for the generative artificial intelligence industry:
- 1. Ensure data and algorithm model security and compliance
- Strictly abide by relevant laws, regulations and policies, collect, use and process data legally and in compliance with regulations, use basic models from legitimate sources, and ensure that the sources of data and algorithm models are legal and their purposes are clear.
- Take advanced technical means and management measures to ensure data security, prevent data leakage and data abuse. Ensure the fairness and impartiality of the model and prevent algorithmic discrimination.
- 2. Promote the construction of content ecology
- Actively guide artificial intelligence technologies, applications and services to generate positive, beneficial, creative and culturally rich content, deliver positive energy to society and enrich the public's spiritual life.
- Take technical measures to prevent generative artificial intelligence products from generating vulgar, violent, terrorist, obscene and other objectionable content, and maintain a healthy and clear network environment and content ecology.
- Strengthen intellectual property protection of content generated by AI, respect originality, encourage innovation, and promote the prosperous development of content creation.
- 3. Pursue technological innovation and quality improvement
- Continue to increase investment in the research and development of generative artificial intelligence technology, promote technological innovation and breakthroughs, and continuously improve the security, reliability, controllability and fairness of artificial intelligence technology.
- Establish and improve a quality assessment system and supervision mechanism, continue to strengthen research and development and investment in content review technology, and strictly review and control AI-generated content to ensure that it complies with policies, regulations, standards and specifications.
- IV. Follow Values and Ethical Standards
- Respect human dignity, rights and freedoms, and take technical measures to prevent the application of generative artificial intelligence technology from infringing on the privacy of others and preventing the generation of discriminatory and negative value information.
- Adhere to value alignment and ensure that technologies and products follow the core socialist values and comply with ethics and social expectations.
- Follow the principles of fairness, equity and transparency, add labels to content generated by AI in accordance with relevant requirements, do not use generative AI technology to manipulate public opinion, standardize the application of generative AI technology, and maintain the fairness of the public opinion environment and social fairness and justice.
- V. Promoting exchanges, cooperation, and openness and co-construction
- Actively cooperate with academia, business, government departments and other parties to jointly explore the application scenarios and development directions of generative artificial intelligence technology and promote the coordinated development of the industry.
- Adhering to the concept of openness and co-construction, we will promote the exchange and sharing of generative artificial intelligence technologies and achievements, contribute to the healthy ecology and coordinated progress of the industry, and contribute to the healthy development of the artificial intelligence industry.
- We are well aware that self-discipline is the essence of industry development and the key to winning social trust. We will take this self-discipline initiative as a guideline, consciously accept supervision from all sectors of society, continuously improve ourselves, contribute to the development and prosperity of the generative artificial intelligence industry, and create greater value for the progress of human society!