AIGC should also be well governed, and the media industry will explore its new scenarios and new spaces

As the new round of technological revolution continues to break through the boundaries of cognition, how can the media industry use AIGC technology to explore development space, and how will it deal with the ethical risks it may bring?

As one of the TV forums of the 29th Shanghai TV Festival, on June 27, a "Focus on the New Quality Productivity of Media - AIGC Use and Governance Dialogue" was held at Shanghai Radio and Television Station. Nearly 100 domestic and foreign industry and academic experts participated in the dialogue to discuss the use and governance of AIGC by the media, as well as the application and development of generative AI in the radio and television industry. AI simultaneous interpretation technology was used at the forum, and Chinese and foreign guests could see the translation of Chinese and English in real time through the large screen.

At the meeting, the "International Intelligent Communication Governance and Strategy Research Center" jointly established by Shanghai Radio and Television Station and Shanghai University was officially unveiled. At the same time, the "Global AI Media Development Research Report (2023-2024)" (pre-release), the "Generative AI Credibility Index System Report" (pre-release), and the CBN Media Model (pre-release version) were released at the same time.

Fang Shizhong, Deputy Director of the Propaganda Department of the Municipal Party Committee and Secretary of the Party Committee of Shanghai Radio and Television Station, introduced that in the face of the opportunities and challenges brought by new technology, Shanghai Radio and Television Station has accelerated the application and expansion of AIGC in the media field, laid out the main direction of "SMG in AI", and is committed to reshaping the media technology base with AI and promoting the development of intelligence, ultra-clean and mobile.

Duan Yong, deputy secretary of the Party Committee of Shanghai University, believes that the technological revolution brought about by AIGC has created many new opportunities for the development of new quality productivity in the media industry. At the same time, the structural flaws of artificial intelligence and the algorithmic bias of AIGC also determine that news fraud and communication ethics cannot be solved by purely technical problems.

"Whether the emergence and evolution of AIGC is Aladdin's magic lamp or Pandora's box requires hot action and cold thinking." Fang Shizhong believes that the issues of news authenticity, copyright protection, data security, ethics and other issues that may be brought about by AIGC are also worthy of attention, and hopes that with the help of this forum, the media and academic circles can reach a broad consensus on the use and governance of AIGC in the media field.

Technology and media empower each other

Focusing on the application status, challenges and future trends of AIGC in the media industry, guests from different fields such as the radio and television industry and technology companies shared their latest observations and thoughts.

Chen Jiangjiang, director of the application and promotion platform of the State Key Laboratory of China Central Radio and Television, explained in detail the "White Paper on the Development of Artificial Intelligence of China Central Station" on the spot, and introduced the exploration of the new generation of AIGC technology in media business forms and scenarios. In his view, AI applications in the media field should follow the principles of co-creation, sharing, and co-construction, advocate the openness of the technology partner ecosystem, promote the optimization of technical products and services and prices through competition, and avoid monopoly. He believes that the use of AI technology needs to be restrained and rational, and large models have the characteristics of a double-edged sword.

Song Jiongming, Deputy Secretary of the Party Committee, Director and President of Shanghai Radio and Television Station and Shanghai Culture Radio, Film and Television Group Co., Ltd., introduced the innovation and responsibility of Shanghai Radio and Television Station under the wave of AIGC. Song Jiongming introduced that as early as 2017, Shanghai Radio and Television Station had initially established an AI laboratory, and after years of development, it has achieved good results in high-definition video restoration, virtual digital human and other application fields; This year, it took the lead in establishing a generative artificial intelligence media integration innovation studio in the field of radio and television in the country, and established six main directions, including the exclusive AI model of financial media and the application of news and information large models, to encourage cross-departmental and field collaboration and deep integration of different types of work.

According to reports, with the blessing of AIGC technology and integration tools, Shanghai Radio and Television Station has continuously upgraded the intelligent level of content production, and has released a series of AIGC public service announcements so far; Jointly developed an interactive digital human sign language interpretation public welfare project based on a large model of sign language with ShanghaiTech University, serving more than 27 million hearing-impaired people across the country. This year, it will also combine AI algorithm matching technology to launch a marriage and love program to explore the possibility of combining AIGC with more program formats.

In view of the risk of AIGC use and security governance, Song Jiongming believes that AIGC should be "well used" and "well governed", and the principles and scales of human values in the technical system should be clarified, such as clarifying that when using artificial intelligence for content production, safe and credible tools must be used to ensure that the generated content products comply with national laws and regulations, industry standards and moral and ethical requirements. Establish strict management measures to ensure that the data collection, storage, processing, use, sharing and other links involved in the application of AIGC technology comply with the requirements of laws and regulations, and unauthorized data access, leakage, tampering or abuse are strictly prohibited.

Managers of digital technology companies shared their current explorations and applications in the field of AI. In their speeches, they all mentioned similar views, that the technological innovation of technology companies and the content creation of the cultural media industry can achieve more efficient integration and development with the blessing of AI.

Ye Hanghui, general manager of Baidu's intelligent cloud media business department, introduced that in the era of large models, Baidu intelligent cloud will inject new intelligent kinetic energy into the media industry from the four dimensions of creativity, reach, appeal and productivity. Li Xingye, vice president of the digital entertainment division of SenseTime, introduced SenseTime's large model matrix and virtual digital human, which has previously cooperated with Shanghai Radio and Television Station to identify the movement trajectory through AIGC technology during the broadcast of sports events, bringing a better viewing experience to the audience, and will carry out more cooperation with TV stations at this year's Paris Olympic Games, and will pay more attention to the creation of human-machine collaboration in the future.

Shu Wenqi, General Manager of Tencent Cloud Media and Neo-Cultural and Creative Industry, shared the application of AIGC in Tencent Meetings, News, Video, Advertising and Operations, and said that for the media industry, it will combine artificial intelligence technology to provide solutions in content production and creation, advertising delivery and analysis in the future.

Academic experts shared the latest research results on AI governance around the world. Eric Poleyn, a professor at Chalmers University of Technology in Sweden, introduced "AIGC Use and Governance: European Cases and Experiences", and interpreted the world's first AI law, the European Union's "Artificial Intelligence Act", to explore value- and ethics-based risk assessment methods, and analyze the trade-offs between different policy options and governance measures. Professor Kim Sung-teol of Korea University introduced the application of generative AI in the production of Korean creative industries in detail through rich cases, discussed its problems at the legal, ethical and socio-economic levels, and reflected on the harmonious coexistence of humans and AI in the future.

The latest results are released

At the dialogue, guests from industry, enterprises and academia focused on AI-generated content, media applications and ethical issues from different perspectives, combined with AIGC's technical empowerment of the radio and television industry, and discussed content production, media ecology, business applications, security challenges, and the impact on higher education and academic research.

Wang Hu, associate dean and professor of the School of Journalism and Communication of Shanghai University, released the "Global Artificial Intelligence Media Development Research Report (2023-2024)" (early edition), sharing the global development and application of artificial intelligence technology, including Canada's combination of research resources and commercialization based on industry, academia and research, the United Kingdom's acceleration of the transformation to an artificial intelligence economy, Germany's creation of an artificial intelligence manufacturing brand, Sweden's focus on traditional advantageous industries, and Israel's focus on intelligent transportation, health and smart cities.

Wang Hu believes that there is a high demand, low retention and low stickiness in the development of AIGC: "What AIGC lacks is not user scenarios and customer needs, but how to develop more new scenarios and new services to create deep value for users, so that they can retain and convert into active users." ”

Yu Weihua, a professor at the School of Media and Communication at Shanghai Jiao Tong University, released the "Generative AI Content Credibility Evaluation Index System Report" (early version) using ten indicators such as accuracy, completeness, timeliness, and unbiasedness. He mentioned that "usable" AI and "trusted" AI are two concepts, and the future AI world must be based on trustworthiness. In his view, the governance of generative AI should return to scientific communication, rather than being confused by marketing: "Generative AI is not as good as imagined, large models are not as strong as imagined, there are still many risks to data sources, and evaluation is not to limit development, but to make it develop in the field of rational science." ”

Yang Yudong, editor-in-chief of Shanghai Radio and Television Station Yicai, introduced Yicai's exploration and thinking on AIGC. In his view, based on cross-border attributes, financial media in the AI era has broad market prospects and development opportunities, such as the current technology in the intelligent generation and intelligent interpretation of professional analysis of financial reports has become mature. At the 2024 World Artificial Intelligence Conference to be held on July 6, the CBN media model will also be officially released.

According to reports, this large model is an AI agent that injects high-quality financial media data and professional domain knowledge on the basis of the basic general corpus to serve the production of financial financial media content, and it is expected to become a financial think tank and assistant in the future, providing investors with first-class financial information services in the rapidly changing dynamic process of the global capital market.

(This article is from Yicai)